ii

Degree Project at Master Level

Comparative study on user’s perception of privacy issues on Facebook: Young adults vs. adults in Sweden

Author: Irma Obradovac & Juan T. Rodriguez Supervisor: Jennifer Contreras Coloma Examiner: Anita Mirijamdotter Academic term: HT17 Subject: Informatics Level: Master Course Code: 4IK50E

iii

Abstract Facebook is a social network that has more than 1.5 billion active members. With the term ‘active’ we mean users who sign in at least once every day on Facebook. Sharing personal information has become a popular activity within the development of all the social networks. Facebook, for instance, contains settings for privacy that allow the user to adjust how "far" the user wants to share his/her own personal information. The problem comes when this can lead to private issues and concerns. The aim of this research was to study the privacy issues perceptions of people within two different ranges of ages, under and over 35 years old. Firstly, we analysed how much people know about privacy on Facebook, and then we tried to understand if there are similarities or differences between those two groups. We thought it would be interesting to know the current state of this phenomena and there was no much research done in Sweden about it. The research was conducted using the mixed method because we wanted to get more global responses from a larger number of respondents in a survey and we wanted to get more detailed answers from an open questionnaire through interviews. We intended to know how is the general knowledge about privacy on Facebook among Swedes. The results showed that people who belonged to the adults group took slightly more care on their privacy than the young-adults group. Also, we discovered that, in general, people are more informed about the risks coming from privacy leaks than a few years ago.

Keywords Social media, Social Network Site, Facebook, personal information, private information, privacy, sharing information, age, old people, young people, old adults, young adults

iv

Acknowledgement Firstly, we would like to thank our supervisor Mrs. Jennifer Contreras Coloma for helping and guiding us to write this thesis. She answered our emails very quickly and gave us invaluable feedback in every step during the research process. We truly appreciate her professionalism as much as her dedication to her work. Then, we want to also thank our examiner Prof. Anita Mirijamdotter and the module leader Mr. Sarfraz Iqbal. Anita and Sarfraz provided us with several educational and interesting seminars as well as constructive criticism and useful feedback for our thesis. Moreover, we are grateful to the rest of professors from the Faculty of Informatics at Linnaeus University whose courses provided us with the necessary knowledge to write this master thesis. Furthermore, special thanks go to all the respondents who helped us to reach the results of our research. Thanks for the time they dedicated to us as well as their responses. They allowed us to conduct this research to the end and their responses gave us important findings that extended our understanding of the concept of privacy on Facebook. In addition, we thank our friends, colleagues and families who supported and encouraged us all the way throughout our research paths. Irma Obradovac wants to thank her boyfriend Goran from the heart for all his love, support and understanding to her "madness" while writing this master thesis. Finally, we want to express our appreciation to the two researchers in this thesis work. During the time we worked together we enjoyed our “ups and downs” and we learnt a lot from each other. There were discussions, arguments and disagreements that, at the end, just helped to improve this work. We understood throughout this time that this boat that was our thesis required that we rowed in the same direction if we wanted to reach our common goal.

v

Table of Contents

Abstract _____________________________________________________ iii

Acknowledgement ____________________________________________ iv

Figures, Tables and List of Abbreviations __________________________ vii

1. Introduction _________________________________________________ 11.1. Research Setting and Motivation _____________________________________ 21.2 Purpose Statement and Research Questions _____________________________ 31.3 Topic Justification _________________________________________________ 31.4 Scope and Limitations ______________________________________________ 51.5 Thesis Organization _______________________________________________ 5

2. Review of the Literature _______________________________________ 72.1 Facebook ________________________________________________________ 92.2 Privacy ________________________________________________________ 102.3 Privacy on Facebook ______________________________________________ 11

3 Methodology _______________________________________________ 143.1. Methodological Tradition _________________________________________ 143.2. Methodological Approach _________________________________________ 173.3. Data Collection _________________________________________________ 20

3.3.1 Collection of quantitative data ___________________________________ 203.3.2 Collection of qualitative data ____________________________________ 21

3.4. Data Analysis ___________________________________________________ 223.4.1. Analysis of quantitative data ____________________________________ 223.4.2 Analysis of the Qualitative Data _________________________________ 22

3.5 Reliability, Validity or similar ______________________________________ 233.6 Ethical Considerations ___________________________________________ 24

4. Empirical Findings __________________________________________ 254.1 Evidences from the quantitative part _________________________________ 254.2 Evidences from the qualitative part __________________________________ 31

5. Discussion _________________________________________________ 355.1 Findings Discussion ______________________________________________ 35

5.1.1 Discussion of Research Question _________________________________ 355.1.1.1. Quantitative discussion ____________________________________ 355.1.1.2. Qualitative discussion _____________________________________ 375.1.1.3. Comparison of quantitative and qualitative discussion ____________ 385.1.1.4. Answer and conclusion of Research Question ___________________ 38

5.2. Overall Discussion _______________________________________________ 40

6. Conclusion ________________________________________________ 426.1. Contribution ____________________________________________________ 426.2. Future Work ____________________________________________________ 43

vi

7. References _________________________________________________ 44

Appendices __________________________________________________ 50Appendix A: _______________________________________________________ 50Appendix B: _______________________________________________________ 52Appendix C ________________________________________________________ 54Appendix E ________________________________________________________ 56

vii

Figures, Tables and List of Abbreviations Figure 1: Master thesis organisation……………………………………………………. 5 Figure 2: Privacy Settings and Tools View………………………………………...…. 12 Figure 3: Three Paradigms Explored…………………………………………………...15 Figure 4: Paradigm Decision Tree……………………………………………………...16 Figure 5: Concurrent Triangulation Design………………………………………...…. 20 Figure 6: Histogram that represents the age of the respondents in absolute numbers.…24 Figure 7: Comparison of the groups’ privacy levels on Facebook……………………..26 Figure 8: Opinions about sharing too much private information on Facebook………...27 Figure 9: Comparison of losing your privacy by using Facebook……………………..28 Figure 10: Group 2 shared this private information more publicly than group 1………35 Tables Table 1: Strengths and weaknesses of Mixed Methods………………………………...18 Table 2: Percentage distribution of the two groups………………………………….…25 Table 3: Percentage distribution of the two groups (only Facebook users) …………...25 Table 4: Users privacy levels on Facebook and their percentage……………………...25 Table 5: Comparison of the groups’ posts privacy levels on Facebook………………..25 Table 6: Respondents who have read the privacy policies on Facebook………………26 Table 7: Comparison of understanding the main concept of privacy on Facebook……27 Table 8: Comparison of changing the privacy settings on Facebook…………………..27 Table 9: Comparison of sharing affects privacy on Facebook…………………………28 Table 10: Comparison of losing your privacy by using Facebook……………………..28 Table 11: Different privacy settings for private information within the group 1………29 Table 12: Different privacy settings for private information within the group 2………29 Table 13: Concepts of Age, Period and Frequency…………………………………….30 Table 14: Concepts related to Privacy within group 1…………………………………31 Table 15: Concepts related to Privacy within group 2…………………………………32 List of Abbreviations ICT Information and Communication Technologies IS Information System SNSs Social Networking Sites

1

1. Introduction ______________________________________________________________________ This chapter introduces the research study by presenting a background of the field of study and presents the reader about the thesis problem. Then, the research settings and the motivation why this topic is interesting to do research are presented too. Furthermore, we also state the purpose statement and the research questions that will guide all the work. Finally, a justification of the topic, the scope and limitations, and how this thesis are organised are shown. ______________________________________________________________________ The World Wide Web (WWW) and its emergence in the late 90s facilitated a new world to individuals, organizations and society in general. Many online applications developed using the WWW’s potentials, for instance, the social networking sites (SNSs), which attracted billions of users and enabled location-independent social relationship among them. SNSs are universal, widespread and popular forms of communication, entertainment and socialization (Lenhart et al., 2007). There exist a variety of them covering countries from all over the world. Some of these sites are primarily operating in specific countries, like VKontakte (VK) in Russia, QZone in China, and the most popular that are available almost everywhere are Facebook, Youtube and Twitter. These popular SNSs have approximately 2,000 million (Facebook), 1,000 million (Youtube) and 300 million (Twitter) of active users on their sites (Statista, 2017). The number of active users on Facebook emphasizes its popularity. This results in a very big mash up, combining together all types of individuals and groups of people with alike and varied interests. Their users find advantages with the use of these sites such as communicating and improving personal connections, both with friends already known offline in the “real life” and with people known only on the Internet. However, SNSs, as pervasive technology, deeply access their users’ everyday life and tend to become invisible once they are ubiquitous, widely adopted, and taken for granted. SNSs reveal drawbacks when it comes to privacy, and, particularly on Facebook, it represents a serious concern. Nevertheless, Facebook has made student life almost unthinkable without the use of its platform. Since its creation in 2004, this SNS has become both a mirror of social interaction and a basic tool for personal identity and network building among students. Facebook offers attractive facilities for communication and interaction despite of it can also threat the privacy of their users. Facebook’s founder, Mark Zuckerberg, at the Crunchie awards in San Francisco in 2010 stated: “The rise of social networking online means that people no longer have an expectation of privacy.” He continued: “the privacy is no longer a social norm." (Theguardian, 2010). Zuckerberg considered that people do not just feel comfortable when they share information on Facebook, but they are more open to other people. On one hand, young people have gained a reputation for not caring so much about their privacy on social media because of their self-revealing presence. But this can be easily explained by the idea that young people simply have another concept of what privacy involves. According to Moore (2008), privacy is generally defined as the right to control personal information. In despite of Facebook started to provide users with choices

2

regarding different levels of privacy some years ago, a study by Liu, et al. (2011) concluded that those privacy options did not match user expectations. Kirkpatrick (2010, p. 202), in the chapter called “Privacy” declared: “The older you are, the more likely you are to find Facebook’s exposure of personal information intrusive and excessive.” The author wrote this statement in order to explain why Facebook was making decisions that might not be sensible to adult participants. On the other hand, embedded also in Kirkpatrick’s explanation, there is an extensive belief that today’s youth do not care about privacy and do not take action to protect it. In her study, Pirli (2015), she found that users over the age of 50 are still behind other age groups in terms of social media use. However, this is changing and more and more are joining to the social media fashion. Several researches also show this: Madden (2010) wrote that the number of social media users older than 50 doubled between 2009 and 2010; and Kietzmann et al. (2011) reported that users of age 55 and older are the fastest growing demographic on Facebook. The increase of people of all ages using Facebook also brings a problematic related to privacy awareness, privacy risks, and consequences of those two. Significant questions appear here that need to be answered: how do young-adults and old-adults perceive their privacy on Facebook? Is the privacy consideration the same for all people who use Facebook? How user’s behaviour on Facebook can be reflected on their real life’s privacy? Giving people the knowledge and tools of how they can protect themselves from relevant risks is a step forward in preventing future problems and consequences derived from their privacy exposure on Facebook. With this study, we aim to investigate the differences in privacy concerns between young-adults and old-adults in Sweden. 1.1. Research Setting and Motivation The purpose of this thesis is to research how much familiar are people with the term privacy on the social network Facebook and how much they care about the information they share on their Facebook profiles. The thesis is conducted using the mixed method approach and the interpretive paradigm. The methods we used in order to collect the data were interviews on a sample of 13 people and surveys on a sample of 128 people. The 12 people who we interviewed were friends and people related to our current jobs. Out of 128 respondents who filled our surveys, approximately half of them did it on paper in different locations in Växjö (the library, the university, Växjö city) and the other half online filling our Google doc survey that we spread in different groups on Facebook. We conducted a comparative study between two groups of Facebook users based on their age. It can be assumed that students have different social needs than the working population, so have different ideas on privacy and self-disclosure. Self-disclosure “reflects the amount of information shared on a user’s profile as well as in the process of communication with others.” (Krasnova and Veltri, 2010, p.2). The first group represents the young-adult population who are between 16 and 35 years old. The second group, the adults, represents the population with age between 36 and 70 years old. The age of 35 was not a random decision. In their study, Malik et al. (2016, p.462), realized that there was “an extremely significant difference in privacy attitudes among people under and over 35 years of age”.

3

A comparative approach between these two groups was chosen because we wanted to get answers about how the difference of age affects to the different behaviours on Facebook, how much and which group knows more about privacy, how much they care, etc. The idea behind using the method of comparison is that we can get a more detailed insight about the concept of privacy on Facebook. The reason why we decided to focus just on Facebook it is because it has most users, both in the world and in Sweden. The original purpose of Facebook was just sharing information among students at Harvard University, but soon Facebook expanded globally. This original purpose stays even today although its functionalities have significantly expanded. We believe that our research about privacy issues on Facebook between the different age groups of users can contribute to our community in the way that social network users get better acquainted with the concept of privacy. People put curtains in their homes mostly to avoid that passers-by or neighbours see what is happening in their home. However, people publish on Facebook photographs of their homes, the food they eat, the location where they are, the persons they spend time with, how they feel, etc. The question here is why people have a different concept of privacy in reality and on Facebook. Therefore, we consider that our research about Facebook's privacy between the different age groups can give interesting results. 1.2 Purpose Statement and Research Questions The main purpose of this work is to study how people perceive privacy within two different ranges of ages. We also aim to improve users’ privacy awareness in SNSs. We focus on Facebook and the privacy risks that arise with the use of it. Firstly, we analysed how much people know about privacy on Facebook. Then we tried to understand if there were similarities or differences within the use of this SNS among the users in different ages grouped by our two groups. Finally, we were able to answer how their behaviour on Facebook was reflected on their privacy. Based on this, the main research question that will lead this research is: RQ1) How do young adults and adults care about their privacy on Facebook? Then, we have added two more sub-questions that will be answered after we compare the results of the comparison study: SQ1) Is the privacy consideration the same within these two groups who use Facebook? SQ2) Do the privacy risks on Facebook affect the use of this SNS among the two groups? 1.3 Topic Justification Understanding why young-adults and old-adults show more or less concern on Facebook can be very useful for legislators, policy makers, Internet entrepreneurs, and for society

4

in general. Considering that SNSs are being used more and more by adults of all ages (Perrin, 2015), and given that recent studies suggesting that young adults differ from older adults in terms of how they use SNSs (Brandtzæg et al., 2010; McAndrew & Jeong, 2012; Chang et al., 2015), researching about age differences in relation to privacy is an important subject of investigation.

We have found similar studies about age differences and privacy attitudes on Facebook conducted in Turkey, Kezer et al. (2016); in Finland, Malik, et al. (2016); and two in The Netherlands, the first one by Van den Broeck et al. (2015), and the second one by Steijn, W. M. P. et al. (2016). All of them conducted their researches using a quantitative approach. Kezer et al. (2016) found out in Turkey that older adults were less likely to disclose information and they were also less likely to use privacy protection on Facebook. The authors also discovered that frequency of use of Facebook declined with age and older age groups had smaller networks and were less probable to use Facebook for socialization purposes. The gender of the user can be also a determining factor regarding privacy concerns on Facebook. Malik et al. (2016) discovered in Finland that women and young Facebook users were considerably more worried about the privacy of their shared pictures. On the other hand, the older age group was less active using the site, in sharing pictures, and in taking privacy-related protective actions. In the Netherlands, Van den Broeck et al. (2015) investigated Facebook use, privacy concern, and the application of privacy settings on Facebook within three stages of adulthood: emerging (until 18 years), young (18-40 years old), and middle adulthood (40-65 years old). They conducted a quantitative research and found out clear differences between the three adult age groups. Emerging adults were less concerned about their privacy and had a pragmatic approach to online self-disclosure and how they controlled their online privacy restrictions. Older age groups, despite their higher privacy concern, were less knowledgeable of privacy settings on Facebook, and used privacy tools and settings less frequent than emerging adults. Also in the Netherlands, Steijn, W.M.P. et al. (2016) confirmed with their study that older individuals were more probable to associate situations related to personal information and reported more concern regarding their privacy. In his paper, Selvan (2012) wrote some experiences about Dan Goldman, legal counsel for the Mayo Clinic, and his problems at his organization due to patient privacy violations in SNS. Goldman once said: “Understanding how employees of different ages use social media is also important” (Selvan, 2012, p.1.). He called them “lifecasters”, people to those who had spent too much time broadcasting their experiences online. In Sweden we found one research conducted by Anderchen and Charvát (2016) using a qualitative approach and focused only on university students. We considered that it would be interesting to know also the perceptions of the rest of population who was not part of the university about Facebook privacy issues.

5

What was also innovative in our research was the methodology we used. No previous research had been conducted using a mixed method approach related with the topic in Sweden. In our case, this meant a number of advantages such as: we could answer a broader and more complete range of research questions because we were not confined to a single method or approach; we could use the strengths of an additional method to overcome the weakness of another method; we could provide stronger evidence for a conclusion through the convergence and corroboration of the findings; we could add insights and understandings that could have been missed when just a single method is used; and we could used the mixed methods approach to increase the generalizability of the results. Therefore, our work can help to affirm or contradict the existing knowledge about the phenomenon, and so complete what is already established. 1.4 Scope and Limitations This work was intended to be the final project within the one-year master program in Information Systems. We had roughly four months to finish it. This tight time-span was a limitation in terms of an ideal longer period for data collection and analysis of the results. One of our challenges was to get acquainted with the different methods and methodologies and the choice of those that gave us the best results of our research. Even though, our goal with this thesis is to increase the knowledge in the fields of information systems, social media and information privacy. Due to we are two people conducting this study we chose to share a mixed methodology in order to get a larger amount of data. Within this methodology, we conducted the qualitative part through the research method of interview and for the quantitative part we used surveys. We believe that the amount of data we collected from these two methods mentioned before, although it is not ideal, it can be enough to help us to answer the research questions of this research. The lack of previous experience with interviewing people and doing surveys can affect in some way our results. Nevertheless, we got all the help we needed to complete this work. For this study, only people living in Sweden with a Facebook account were taken into consideration. For the quantitative part, a survey was distributed both in paper and online. In the time-span of three weeks we collected 128 answers. For the qualitative part, we have successfully conducted 13 interviews, face to face, among our friends and people related to our current jobs who live in Växjö and in Ljungby. In order to achieve more accurate results, one of our challenges was to find equal number of participants for both of our age groups on Facebook. Anyway, facing the problems and challenges that we encountered during this research work has encouraged us to be more ambitious and more capable in confronting them. 1.5 Thesis Organization This thesis consists of six main chapters with their additional sub-chapters. The first chapter presents an introductory review for this work where the reader gets acquainted key concepts of our research. It starts with an introduction containing all the related work we found within the field of social media, Facebook and privacy issues. Also, we presented here our research settings and the motivation that leaded us to become fascinated with the topic. We continued the chapter presenting the purpose statement and

6

the research questions that led this research. Finally, we justified the topic and the lack of research in our context and we explained the scope and limitations. In the second chapter we discuss the significant literature review about our field of study, having a subchapter for Facebook, privacy on Facebook and SNSs privacy. The third chapter shows the methodology we used in this study as well as how we managed the information in order to be reliable and valid. We close the chapter with the ethical considerations we put in practice within all the work. The chapter four shows and discusses all the empirical findings we got from both the interviews and the surveys. In the chapter five we discussed the results and in the chapter six we write about our conclusions, our contribution to the field and we suggest future research.

Chapter 1: Introduction

Chapter 2: Review of the literature

Chapter 3: Methodology

Chapter 4: Empirical Findings

Chapter 5: Discussion

Chapter 6: Conclusion

Figure 1: Organisation of this master thesis. Source: authors

7

2. Review of the Literature ______________________________________________________________________ This second chapter presents the literature review and the theory used in our research study. The section firstly describes the literature work around the field of study and later on it explains more in detail concepts that we think are necessary to define for a correct comprehension of this master thesis, such as, Facebook, the concept of privacy and the privacy applied to the SNS. ______________________________________________________________________ As humans, we have always engaged in social networks. They can be formed around our family, friendships, religious beliefs, social status and more. Social media is any kind of information we share with our social network using social networking websites or services. Examples are blogs, forums, photos, audio, profiles and status updates (Eley & Tilley, 2009 p. 78). Social media allows people with no special knowledge to post and share content with the world instantly. Last few years social networking has become increasingly popular with hundreds of millions of users worldwide. Current big portals for social media sharing are Facebook, Twitter and YouTube (Eley & Tilley, 2009, p. 78). While these social networks may have started up as the typical geek hangout, just like the Internet did in the 90s, it has now become mainstream. Everyone, from college students to business people to celebrities to politicians has embraced this new channel (Eley & Tilley, 2009 p. 79). Social media is even thought to have played a major role in helping the US President Barack Obama win the election in 2008 (News Blaze, 2009). Social media is a developing area and a current buzzword, both for companies and in the academic world. It is already huge amongst private persons. Facebook now has over 1.86 billion monthly active users world wide (Zephoria, 2017) and Twitter more than 310 million accounts (Statista, 2017). With this amount of billion users, the decisions that Facebook takes about its privacy settings have the impact to influence many people. There have been times when these decisions have caused controversy; even so, Facebook has succeeded to attract more and more users to its service. In 2010, Facebook’s founder Mark Zuckerberg affirmed: “People have really gotten comfortable not only sharing more information and different kinds, but more openly and with more people. That social norm is just something that has evolved over time.” (Kirkpatrick, 2010). This statement came after Facebook’s move in December 2009 asking users to reconsider their privacy settings (Zuckerberg, 2009). In the chapter called “Privacy”, Kirkpatrick (2010) stated: “The older you are, the more likely you are to find Facebook’s exposure of personal information intrusive and excessive.” The author explained with this declaration why Facebook, which has always been popular with young people, was taking decisions that might not seem sensible to adult users. Lipford et al. (2008) affirmed that many privacy mechanisms in SNS are weak in purpose in order to facilitate joining the platform and start sharing information. Users are not so much aware about the existing privacy mechanisms and research has shown that the reasons of this include poor interface design, permissive default settings, social conformity and a natural trust in the online community (Lipford et al., 2008).

8

In their paper: “Facebook and Online Privacy: Attitudes, Behaviors, and Unintended Consequences”, Debatin et al. (2009) concluded that the gratifications of using Facebook tend to compensate the perceived threats to privacy. They added that decreasing profile visibility through restricting access to friends is just a quick fix and not a systematic approach to protecting privacy. Most users do not realize that restricting visibility to their data does not protect them from the risks resulting from the amount, persistence and quality of the data they provide. The common privacy controls include limitation of profile access, blocking and hiding other site users, and item-level access control. The conceptualization of these privacy controls and privacy behaviours vary among the SNS. Stutzman and Kramer-Duffield (2010) explored the quite popular behaviour of setting one’s Facebook profile as friends-only. A friends-only profile means that the profile is private and is only visible by your connections. The authors found that “gender, network size, weak tie expectancy violations and increasing levels of interpersonal privacy practices” were linked to having a friends-only Facebook profile (Stutzman and Kramer-Duffield, 2010, p.1561). They added that although privacy in SNS is managed by a variety of methods, the act of setting a friends-only profile is exceptionally significant. The users who have done this show a common connection identity to the network, limiting contribution to and resources drawn from the including network. In their paper, Boyd and Hargittai (2010) examined the attitudes and practices of a group of young people (between 18-19 years old) during 2009 and 2010 about Facebook’s privacy settings. They found that, during that year, modifications to privacy settings had increased, and both frequency and type of Facebook use in addition to Internet skills were correlated with making those modifications to privacy settings. However, they perceived few gender differences in how the group of teenagers approached their Facebook privacy settings. Liu et al. (2011) wrote an article called “Analyzing Facebook privacy settings: User expectations vs. Reality” where they concluded that privacy, in general, is hard to measure, particularly since it is hard to quantify even by the users themselves. Pictures, for instance, are likely to have very different privacy requirements, depending on who is in the picture, where it was taken, etc. Furthermore, added that when they measured the users’ ideal privacy settings, they treated the users’ answer as ground truth. However, this might not always be the absolute truth since users’ answers can differ with time or the user’s may have not completely thought the implications of a given setting. In another study by Waters and Ackerman (2011), four motivations for active users of Facebook who share private information were named. The first one was conceiving sharing private information as an enjoyable and fun activity. The second was that users tend to perceive it as a way to store information that can be significant to them or use it as a form of entertainment. The third one was to keep up with trends and the fourth and final motivation was that users share to show off their popularity or announce events to one another. The authors found that one positive consequence of disclosing is that users feel in control of “relationship management/psychological well-being”. A negative consequence, nevertheless, is that users spend too much time on Facebook. Waters and Ackerman concluded that motivations and consequences of disclosing are alike for males and females, but females disclosed mainly for entertainment and to store information, and males disclosed primarily for information sharing.

9

A similar result was presented in the paper called “Privacy management in a connected world: Student’s perception of Facebook privacy settings” by Bornoe and Barkhuus (2011). In their research, they tried to answer the following questions: Why at all discuss privacy when SNS are all about exposing information? And, why does privacy matter at all? They discovered that the participants in their study got different forms of gratifications from using SNS, for instance, they wanted both, to engage in social interactions, and expose information. This does not mean that they gave up privacy, “rather exposing leads to a need for privacy management” (Bornoe and Barkhuus, 2011). In general, they concluded that the participants got to know the media better, which gratifications and benefits they got, and how it was utilizable to their stage in life. They received almost only positive feedback about using SNS and, even though they had some lack of awareness of privacy settings, it was compensated for through the practical use. In his paper, Tufekci (2012) examined a sample of 450 young students and found that they were reacting pro-actively and adjusting their privacy settings. The students did not seem uncaring about their privacy options and were not waiting until they got burned by the impact of negative personal experiences. Negative experiences like the ones described in “From Facebook Regrets to Facebook Privacy Nudges” (Wang et al., 2013): the pierogi mascot of a baseball club who was fired after posting critical comments about his club on his Facebook, or the high school teacher who was also fired because she posted a photo on Facebook in which she was holding a mug of beer and a glass of wine. Wang et al. (2013) showed in their study that people have various cognitive and behavioural prejudices that affect their decision making and they make posts that they later regret. These regretted revelations sometimes lead to substantial consequences like loss of a job or a relationship. Despite of the risks and threats against the user privacy on social networks, these same users were found to ponder more benefits likely and fewer risks likely to happen in comparison with non-users of social networks (Steijn, W.M.P, et al., 2016). Users generally have a positive risk-benefit balance regarding the sharing of personal information while non-users have a negative risk-benefit balance on average. 2.1 Facebook Created in November 2004 by Harvard University undergraduate Mark Zuckerberg, Facebook was envisioned as a forum for information flow and student interaction on college campuses in USA. Facebook has expanded outside the university environment and is now open to the general public. What differentiates Facebook from other SNSs is that the profile accessibility and participation are structured by the user’s offline network, for instance, their university, workplace or city. Moreover, people join to Facebook and create networks having only members of their chosen offline contexts. Because of this, a person outside a user’s network may not see his/her complete profile unless the two become friends on the social network (Strater and Lipford, 2008). According to Debatin et al. (2009, p.1), “this popular social network service has quickly become both a basic tool for and a mirror of social interaction, personal identity, and network building among students”. Users join to Facebook mainly to keep in touch with distant friends and to search for people recently met at some place like an event, a party

10

or workplace. Besides, users assume that these people become the primary audience of their own profiles (Lampe et al., 2006). Facebook profiles can contain a big amount of information, including details or interactions of the user in the social environment, for instance, pictures, friends lists, and messages with friends, and a variety of self-reported information. Strater and Lipford (2008) suggest three categories where users can disclose the information in their profiles:

- Basic Information: Basic descriptive information like birthday, university, and hometown.

- Contact Information: Including phone number, email, and personal websites. - Personal Information: Descriptive information to inform interests, likes, and

personality.

The rest of the user’s profile exposes users’ activities and social features within their social network, such as:

- Photos: Users can upload images, make comments and identify friends in them by using “tags”. A “tagged” friend is able to view the photo in his/her own profile.

- Friends: Users can request and accept friends, who will be then linked through profiles.

- Wall: Friends can write messages, links, pictures or videos that remain posted on the profile for others to read/see.

- Newsfeed: A list with the user interactions on the profile, such as likes, comments, adding a friend, etc. This information is also displayed on friend’s homepages.

2.2 Privacy Privacy is a fundamental human right. This is recognized by the Article 8 of the European Convention on Human Rights that specifies a right to respect for “private and family life, home and correspondence” (Council of Europe, 2017). Likewise, the Charter of Fundamental Rights of the European Union establishes the “respect for private and family life” (Article 7) and adds a specific article on “protection of personal data” (Article 8). Furthermore, we can find in the Article 12 of the Universal Declaration of Human Rights that an individual is protected from “arbitrary interference with his privacy, family, home or correspondence, and attacks upon his honour and reputation” (Universal Declaration of Human Rights, 2017). In relation to the digital world, Danezis et al. (2014) detected that there was a power disparity between data processing entities, the ones that determine what and how data is processed, and the individuals whose lives can be influenced by decisions based on automated data analysis or by failures to effectively protect private information. Therefore, the protection of privacy plays a very important role, especially in the digital world. Many people are often unaware of the data processing and its consequences when using a specific service. At the same time, our society is more and more dependent on the trustworthy functioning of information and communication technologies (ICT). The down side of this progress is that there are unclear responsibilities and lack of transparency for users, and generally missing guarantees of privacy and security features (Danezis et al., 2014).

11

Managing privacy and identity is a face-to-face interaction that is constantly negotiated. Nonetheless, online interactions make the case-by-case decision-making process more difficult. Users hardly interact with each other synchronously what means that the decisions of privacy must be made a priori and explicitly (Strater and Lipford, 2008). Smith et al. (1996) defined information privacy based on five major dimensions: (1) collection of personal information, (2) internal unauthorized secondary use of personal information, (3) external unauthorized secondary use of personal information, (4) errors in personal information, and (5) improper access to personal information. Information privacy differs with social privacy, which tackles user concerns about self-control over their own personal information (Raynes-Goldie, 2010). Internet users are exposing a large amount of personal information as part of their participation in SNSs. This propagation of personal data presents implies a variety of risks for individuals, for example, identity theft, embarrassment, stalking, and blackmail (Lipford et al., 2008). The need for flexible privacy mechanisms to protect user data increases as the participation in online communities growths.

2.3 Privacy on Facebook Approximately half of the users who have access to the Internet are members of some SNS, causing a fundamental shift in the patterns of context exchange over the Web (Liu, et al., 2011). This shift implies that users, instead of just being content consumers, they are now required to be content creators and managers. Today, the user must decide which of his/her friends, group members, and other Facebook users can access the content he/she shares. The user will always face a dilemma between potential privacy risks and the desire to share content and information. Facebook makes sharing information quite uncomplicated and encourages its members to post personal information like where you are or who you are with. Facebook privacy settings provide users of a good deal of flexibility about who is allowed to see their information (Jones and Soltren, 2005, p.6). Privacy settings allow users to set who can see their private information, pictures, comments, etc. They also provide the ability to block individual users as well as the ability to select who can find us. According to a user agreement, a user may ask Facebook not to share information with a third party, even though the request itself is not in the privacy settings (Figure 2).

12

Figure 2. Privacy Settings and Tools View. Source: facebook.com

Facebook itself has recognized that users’ information may be used without consent: a document published with the Province of Ontario’s Office of the Information and Privacy Commissioner stated: “at any point in time and potentially without any notice, information from your profile and logs of your online activities may be used and disclosed in unexpected ways that can affect your privacy” (Cavoukian, 2007). Facebook, and the rest of SNSs, have changed how many people develop and preserve relationships through sharing and posting personal information. The amount and depth of these disclosures have raised many concerns about online privacy. Strater and Lipford (2008) demonstrated in their results the need for mechanisms that could provide awareness of the privacy impact in the users’ daily interactions. These interactions go from disclosing user’s interests, photos, contact information, daily activities, and associations to interactions with other users or groups. The proliferation of these sensitive data represents a diversity of risks, such as: users might face embarrassing situations or blackmailing by revealing inappropriate data unintentionally. They can even put themselves at risk for online or physical attacks, such as identity theft or stalking. While most of users show knowledge of the privacy concerns and available options, research indicates that users’ privacy attitudes do not impact their decisions to disclose sensitive information (Acquisti and Gross, 2006). There exists little research about user-generated explanations for these behaviours, including the problems users find when managing their privacy over a range of personal and sensitive information (Strater and Lipford, 2008). Facebook gives granular control to the users over the availability of every profile feature and contact information. These controls vary between “only me” to “public”. Users are given also the possibility to restrict access to their profiles to “all your friends”. Strater and Lipford (2008) defined three privacy-related aspects of the profile:

- Public Profiles: Profiles accessible to all people in the network. - Private Profiles: Profiles that can only be viewed by user’s friends. - Search Profile: Profile information returned in search results.

13

In their research, Bornoe and Barkhuus (2011), all their participants had been active on Facebook for at least a couple of years when they observed a process correlated with maturing and social changes. Their use of SNSs had changed towards more intensive considerations about posting information they felt comfortable with and what to make public. Also, they were more careful when accepting new “friend”. Mainly social norms from the offline world seemed to be reflected when using SNSs. With each user acting based on his/her respective cultural background and characteristics, a variety of different behaviours develops on SNSs. Actually, not only the user behaviour is influenced by the respective cultural origin, but existing research shows that different cultures also differ when it comes to privacy on SNSs (Krasnova and Veltri, 2010; Tsoi and Chen, 2011; Wang et al., 2011).

14

3 Methodology ______________________________________________________________________ This third chapter describes the methodology used during the research of this thesis. It contains the methodological tradition, the methodological approach and the methods used for the data collection and the data analysis. In every case we describe the options that were available and a proper justification of the choice that we finally selected as the proper one. This chapter closes with the procedures that we followed in order to ensure the validity and reliability of the data and ethical concerns that we respected. ______________________________________________________________________ 3.1. Methodological Tradition A paradigm is a way of understanding and looking at the world and is built up of a collection of values, views, opinions and understandings on what criteria gives worth to something (Hart, 2005, p.213). All research is grounded on a fundamental ideology that establishes how to conduct research in the best way. There have been numerous attempts to categorize Information Systems (IS) research (Hirschlheim and Klein, 1989; Avgerou, 2000). As a consequence, we refer now to three major schools of thought in the IS discipline: the Interpretive, the Positivist, and the Critical. Each of them contributes to different aspects of the IS discipline. This three-fold classification appears to have been widely accepted within the IS research literature (Klein and Myers, 1999; McGrath, 2005; Myers, 2009; Ngwenyama and Lee, 1997; Richardson and Robinson, 2007; Stahl and Brooke, 2008). Interpretive. Klein and Myers (1999) claimed that the primary assumption of interpretive research is that knowledge about a reality is gained through language, consciousness and shared meaning. The use of this paradigm allows researchers to understand a specific phenomenon by interpreting what people say or believe. As people interact with the world, they get and interpret their own meanings about the reality that already exists (Orlikowski and Baroudi, 1991). Every individual can determine his/her own interpretation of a phenomena and still be correct. Moreover, interpretation sometimes involves prejudices, which means that every individual might be stained by his/her past and the conclusions that the researcher achieves can be therefore wrong. It is important that the researcher who uses the interpretive paradigm is aware of this fact and conducts his/her experiment with an open mind (Klein and Myers, 1999). Positivist. The fundamental assumption behind the positivist research paradigm is that reality is objectively given and can be measured independently of the researcher and his/her instruments. The objective is to test a theory (Myers, 1997). According to Orlikowski and Baroudi (1991), positivist research is conducted with formal propositions, uses measurements of quantifiable variables and involves the testing of diverse hypotheses to enrich the understanding of a specific phenomenon. Critical. In information systems, critical research is focused on social issues such as, freedom, social control, power, and values with respect to the development, use, and impact of information technology (Stahl and Brooke, 2008). Critical research continues underrepresented in the IS research literature (Falconer, 2008), causing some researchers to describe it as “a missing paradigm” in IS research (Chen and Hirschheim, 2004; Richardson and Robinson, 2007). Orlikowski and Baroudi (1991) claimed that some of the significant features of the critical research philosophy are a belief in people’s ability

15

to change their social circumstances and material, although this capacity to change is restricted to predominant systems of political, economic, and cultural authority. This belief affirms that contradictions included in existing social forms lead to inequalities, and conflicts, which lead, at the same time, to the emergence of new social forms. Moreover, it is believed that knowledge is based on historical and social practices. Critical research intends to transform these separating and restrictive social conditions. Therefore, the critical research differs from the positivist and interpretative research because these last two “are content to predict or explain the status quo” (Orlikowki and Baroudi, 1991, p.19). Anderson (2013) summarized the three paradigms in the next table:

Paradigm Ontology Epistemology Question Method Positivism Hidden rules govern

teaching and learning process.

Focus on reliable and valid tools to undercover rules.

What works? Quantitative

Interpretive/ constructivist

Reality is created by individuals in groups.

Discover the underlying meaning of events and activities.

Why do you act this way?

Qualitative

Critical Society is rife with inequalities and injustice.

Helping uncover injustice and empowering citizens.

How can I change this situation?

Ideological review. Civil actions.

This thesis aims to research privacy concerns in a SNS such as Facebook so the interpretive research was chosen in order to understand this social phenomenon. Another motivation for this choice was that the interpretive paradigm helps the researcher to understand the full complexity of human sense making, rather than predefining dependent and independent variables (Kaplan and Maxwell, 1994). Here the intention was to get deep insights of the complex issue of privacy concerns in SNS, what made interpretive research approach the most appropriate here. Interpretive researchers make an interpretation of what they hear, see and understand, and these interpretations cannot come from their own contexts, backgrounds, and prior understandings (Creswell, 2009). The interpretivist approach is based on naturalistic approach of data collection such as observations and interviews. Secondary data research is also popular within interpretivism philosophy. The most remarkable variations of interpretivism include symbolic interactionism, phenomenology, realism, hermeneutics, and naturalistic inquiry (Gray, 2004). Phenomenology is a philosophical research tradition whose fundamental aim is to develop a bigger understanding of individuals’ experiences through the consciousness of the experiencer (Giorgi, 2009). The main emphasis is to interpret or to describe human experience as lived by the experiencer in a way that can be utilized as a source of qualitative evidence.

Figure 3: Three Paradigms explored. Adapted from Anderson (2013)

16

One of the main criticisms of mixed methods research is that it is often implemented uncritically by researchers who pay little attention to the paradigmatic differences between methodological approaches (Sale et al., 2002). Even though it is largely acknowledged within the field of mixed methods research that paradigms do not always lead to specific research methods (Johnson and Onwuegbuzie, 2004), it is commonly accepted that certain methods fit better within complementary paradigms. For this reason, Mayoh and Onwuegbuzie (2015) discussed in their paper the underlying paradigmatic compatibility and differences for postpositivist and phenomenological work. Traditionally, methodological purists have explicitly rejected the possibility that qualitative and quantitative research paradigms could coexist within a single research study. Johnson et al. (2010) claimed that by using multiple methods within a single overarching paradigm researchers conducting mixed methods research can negotiate these differences. This approach is based on the notion within the field of mixed methods research that paradigms do not always lead to particular research methods (Johnson and Onwuegbuzie, 2004). Mayoh and Onwuegbuzie (2015, p.95) demonstrated in their paper that “axiological and methodological parallels between phenomenological and quantitative methods allow for the combination of these methods within a single study with a single overarching paradigmatic framework, which is able to complement multiple methods adequately. This will help provide a specific tailored justification for the adoption of phenomenology in combination with alternative methods.” For this thesis work, we adopted the Mix Methods Phenomenological Research (MMPR) what was defined by Mayoh and Onwuegbuzie (2015) as the research that blends phenomenological methods with methods grounded in an alternative paradigm within a single study. This helped us to provide an analysis from multiple viewpoints and a more in-depth analysis of qualitative data. We wanted to understand our respondents’ experiences with privacy on Facebook as they perceived them, described in their own words. Bentz and Shapiro (1998, cited in Groenewald, 2004, p.13) stated that the root of phenomenology is “the intent is to understand the phenomena in their own terms – to provide a description of human experience as it is experienced by the person him/herself”. Therefore, phenomenology and its variant MMPR seemed to fit with the purpose of our study.

IS Paradigms

Positivism

Interpretive

Critical

phenomenology Mix Methods Phenomenological

Research (MMPR)

symbolic interactionism

hermeneutics

realism

naturalistic inquiry

Figure 4: Paradigm Decision Tree. Source: Authors

17

3.2. Methodological Approach Research methods are processes for collecting data on which basis will implement some research or make some business decisions. The research method that we used in this work was mixed methods. Griensven et al. (2014) defined the Mix Methodology Research (MMR) as “the type of research in which a researcher or team of researchers combines elements of qualitative and quantitative research approaches (e.g. use of qualitative and quantitative viewpoints, data collection, analysis, inference techniques) for the broad purposes of breadth and depth of understanding and corroboration”. The reason for using a combination of these two methods was that each of them gives a different type of data. Griensven et al. (2014, p.367) stated that “quantitative research methods are based on the generation and manipulation of numbers using statistical analysis”, while “qualitative research concerns itself with meaning, social context and personal experience”. On this way, the quantitative method provides a mathematical, statistical and numerical analysis of data collected through surveys, questionnaires, etc. On the other side, the qualitative method searches for the answers, produces findings, collects evidence, etc. The qualitative method offered us to set open-ended questions through the interview process. Furthermore, we got participants to use their word in answering. Through the quantitative method, in a short time we picked up data and participants were limited to the selection of the offered answers. Through the qualitative method we got more detailed answers because we, as researchers, had the opportunity to ask how and why. Data from the quantitative method can be represented using graphs and analysed by descriptive statistics and visual analysis of graphical presentations of the measures while qualitative interviews were transcribed word by word (Mengshoel, 2012, p.374). Mengshoel (2012, p.373) gave more arguments for using this method: “quantitative and qualitative methods have different limitations and strengths, and the limitations of one can be compensated by the strengths of the other and integrating multiple concerns in one study yields more nuanced results than examining them in isolation”. According to Griensven et al. (2014, p.368), if we compare with the single method approach then “MMR may be viewed as providing a more complete and deeper understanding of the subject under investigation, and having a greater scope”. Griensven et al. (2014) suggested the reasons why to use MMR:

• Triangulation => the converging of the results of two different and separate

methods. • Complementarity => using of one study to elaborate on the other one study. • Development => using of one study as a base for the following investigation. • Initiation => discovering paradox and contradiction using different methods. • Expansion => investigating different facts of the entity that is under investigation.

Researchers believe that the MMR has a lot of advantages, while their disadvantages are set off each other. We can conclude that through quantitative research we got statistical power and generalization, while quantitative research provided with significance,

18

deepness of research and context. Scientists have argued that the combination of collecting data throws these two methods “is a natural method of everyday problem solving” (Griensven et al., 2014, p.368). The results obtained through MMR are more precisely and more complex than results obtained from only a single method: “the findings of MMR are believed to be more comprehensive and valid than the findings of quantitative and qualitative studies conducted separately” (Mengshoel, 2012, p.373). The main objective of the MMR is the fact that research questions ride the entire study. Therefore, the researchers recommend to clearly asking the research question about which one speaks in research studies. Quantitative and qualitative research methods have a different paradigms base: “Quantitative and qualitative research methods are based on different paradigms that have both ontological (how reality is viewed) and epistemological (how knowledge and truth are viewed) implications” (Mengshoel, 2012, p.373). A quantitative study is based on: “a positivist/post-positivist paradigm, postulating the existence of a relatively stable material reality that exists across different cultures and contexts”, while a qualitative study is based on: “a constructivist or interpretivist paradigm, postulating that there are multiple constructed realities bound to context, time and culture that can be studied by examining people’s experiences and what is going on in social situations” (Mengshoel, 2012, p.373). The mixed method provides help for “highlight the similarities and differences between particular aspects of a phenomenon” (Östlund et al., 2010, p. 370). Johnson and Onwuegbuzie (2004, p.21) explained in their paper “Mixed Methods Research: A Research Paradigm whose time has come” the strengths and weakness of using a mixed research approach in a study. We adapted it in the next Table 1:

Strengths Weaknesses - - Word, pictures, and narrative can be utilized to

add meaning to numbers. - - Numbers can be used to add precision to words,

pictures, and narrative. - - Researchers can generate and test a grounded

theory. - - Can answer a broader and more complete range

of research questions because the researcher is not confined to a single method or approach.

- - Researchers can use the strengths of an additional method to overcome the weaknesses of another method by using both in a research study.

- - Can provide stronger evidence for a conclusion through convergence and corroboration of findings.

- - Can add insights and understanding that can be missed when just a single method is used.

- - Can be used to increase the generalizability of the results.

- - Can be difficult for a single researcher to carry out both qualitative and quantitative research, especially if two or more approaches are expected to be used concurrently.

- - Researcher has to learn about multiple methods and approaches and understand how to mix them appropriately.

- - More time consuming. - - More expensive.

Table 1: Strengths and weaknesses of Mixed Methods. Adapted from Johnson and Onwuegbuzie (2004, p.21)

19

A mixture of qualitative and quantitative approach is recommended to higher the reliability and validity of the results. Because all the previously said, we decided to use mixed methods in our research. Tashakkori and Teddlie (2003, p.223) identified six major mixed methods design strategies:

- - Sequential Explanatory: Characterized by a collection and analysis of quantitative data followed by a collection and analysis of qualitative data. Its purpose is to use the qualitative results to support the explanation and interpretation of the findings of the quantitative study.

- - Sequential Exploratory: Characterized by an initial stage of qualitative data collection and analysis followed by a stage of quantitative data collection and analysis. The main purpose of the sequential exploratory is to explore a phenomenon but can be also useful when developing and testing a new instrument.

- - Sequential Transformative: Characterized by a collection and analysis of either quantitative or qualitative data first and then the results are integrated in the interpretation phase.

- - Concurrent Triangulation: Characterized by two or more methods used to confirm, cross-validate, or corroborate findings within a research. The data collection is concurrent and generally both methods are used to overcome a weakness in using one method with the strengths of another.

- - Concurrent Nested: Characterized by giving priority to one of the methods and guides the project while the other one is embedded or nested. The purpose of concurrent nested is to tackle a different question than the dominant of to seek information from different levels.

- - Concurrent Transformative: Characterized by the use of a theoretical perspective that is reflected in the purpose of the research questions of the study to guide all methodological choices. The purpose of the concurrent transformative is to evaluate a theoretical perspective at different levels of the analysis. In our research the mixed methods design employed a concurrent triangulation model, which gave equal priority to our concurrently collected qualitative (QUAL) and quantitative (QUAN) data. The capitalization refers to each method being a significant contributor, rather than QUAL-quan where the qualitative method would take priority. This research used a mixed methods approach that included semi-structured interviews and surveys. The mixed methods approach provided of a more comprehensive inquiry to answer questions that one method alone could not address. For example, while the quantitative data collection obtained the numbers of how many people in each age-group had read the Privacy Policy in Facebook, it is the qualitative methods that explored the complexity of the question with why and why not people had done it. This helped us to answer the research questions from a broader and deeper perspective. For the study to be truly mixed methods, the two approaches were integrated during the analysis phase (Figure 5).

20

3.3. Data Collection This project is methodologically innovative for using a mixed methods approach in order to try to understand the privacy perception in Facebook by people of different ages. We have to mention that both parts of the study, the quantitative and the qualitative, were conducted in a single phase during April 2017 due to a time limitation. Using both type of data helped us overcome limitations in one method and gave us a better degree of understanding about the research problem, as suggested by Creswell (2009). 3.3.1 Collection of quantitative data The main and only method for collecting quantitative data in this study a survey. Saunders et al. (2016) stated that a survey is a form of a data collection method that allows the researcher to present the participants exactly the same set of questions, and later analyse them with the help of a computer. It is mainly utilized to examine or explain relationships. The principal advantages with using a survey as a data collection method are that it is easy to administer, it provides consistent data and the interpretation and analysis of the information obtained are simple (Malhotra et al., 2010). On the other hand, the disadvantage is that respondents might be unwilling to respond, sometimes as a consequence of that the topics being covered are sensitive (Malhotra et al., 2010). We conducted the quantitative part of the study through a survey containing 14 questions with multiple-choice of answering (See Appendix B). The intention was to deliver this survey to a sample of between 100 – 150 respondents with different ages, education level, gender, and profession. The survey was available in three forms: 1) paper, where we met face-to-face with the participants, 2) online through e-mail and a group message/sharing on Facebook and 3) phone call. Most of the questions in the questionnaire were closed and were measured by five point Likert’s scale, which was measured by asking the respondents how much they agree or disagree with a statement in a degree of one to five points rating scale.

Concurrent Triangulation Design

QUAN QUAL

QUAN Data Collection

QUAL Data Collection

QUAN Data Analysis

QUAL Data Collection Data Results Compared

Figure 5: Concurrent Triangulation Design. Adapted from Creswell (2009).

21

We employed the paper form in the following locations in Växjö: Lexicon school, Hermi Bil and Fortnox. The Lexicon school is a private school providing adult extra education services. The reason for choosing this location was because we found a large number of participants from the second group (36+ years), with different educations and professions. Hermi Bil is a company that deals with car sales and services. We chose this location because this place of sale is visited daily by different clients and it was possible to choose respondents four our survey. Fortnox is a company that deals with the development of administrative software programs and employs different people within our two age groups with different levels of education and profession. The second way to spread our survey was making a Google-doc survey and share it on University groups on Facebook to pick up answers from the younger population. When we realized that we still lacked a certain number of participants in the older group, we decided to target the phone and call a few friends, acquaintances, relatives, etc. who we knew they used Facebook and that their age matched with the ages we needed to our research. In the survey, the participants firstly indicated some basic information about themselves, like age or gender. After that, the participants answered the question if they used Facebook. In order to understand the participants’ perception with privacy on Facebook, we asked them if they had ever read the privacy Concerns on Facebook (yes/no), how they felt about Facebook's Privacy Policies (multiple-choice question), if they shared too much private information (scale 1-5), if they understood the main concept of privacy on Facebook (1-5 scale), etc. Furthermore, the participants specified which type of information (birthday, relationship, about me, photos and videos, etc.) they shared and who with (just me, only friends, everyone, etc.). In all of these questions the participants could only choose one answer. In the last one, the 14th question, we gave the option that participants wrote whatever they wanted about this topic. 3.3.2 Collection of qualitative data The qualitative part of the study was conducted through interviews. We designed a questionnaire that included 12 open-end questions and we had the opportunity to interact face-to-face with the participants (See Appendix A). We tried to design the questions in a way that they would invite the respondents to describe their experiences and explain things from their point of view, as phenomenology requests. The participants provided with their own answers and expressed what they felt about the questions we asked and the topic in general. Also, the participants were voice-recorded with their prior consent, so we could go deeply in their answers lately. We set a sample of 13 respondents for the interviews, half of them were conducted in Swedish and the other half in English. This amount was in accordance with what Creswell (1998, cited in Groenewald, 2004, p. 11) and Boyd (2001, cited in Groenewald, 2004, p. 11) suggested about up to 10 participants being enough for a phenomenological study. However, since our study is a comparison we saw necessary to include a few more in order to get more accurate results. It took an average of 15 minutes to conduct the interviews and around 45 minutes to transcribe verbatim (word by word) them. These transcriptions were used as a sanity check to verify later on back to the respondents that

22

no information was lost from the interviewees’ answers during the sessions. The reason for having chosen these people was that they were our friends and people related to our current jobs what made it relatively easy to get their help. The idea behind these interviews was to collect equal number of people in both age groups because only in this way will we could get an objective answer to our research question and sub-questions. 3.4. Data Analysis 3.4.1. Analysis of quantitative data Quantitative data analysis is useful in evaluation because it gives easy to understand and quantifiable results. Quantitative data can be analysed in a variety of different ways. There are four levels of measurement associated with quantitative data and each one of them can influence the type of analysis one can use. They are: nominal, ordinal, interval and ratio (scale) (Velleman and Wilkinson, 1993). Nominal data: data is basic classification data and has no logical. For example, male or female. Ordinal data: data has a logical order but the disparities between values are not constant. For example, privacy levels (public, just me, all my friends) Interval data: data has a logical order and is continuous. For instance, items measured on a Likert scale (1 = strongly disagree, 2 = somewhat disagree, 3 = neither agree nor disagree, 4 = somewhat agree, 5 = strongly agree). Ratio data: data is ordered, continuous, has standardized differences between values, and a natural zero. For example, age, height, weight. In our study we have all the types of quantitative data mentioned above. In order to answer our research questions, we used the methods of Data Tabulation and Descriptive Statistics. Using Data Tabulation we tabulated the results we got from the surveys for the different variables in our data set. This gave us a comprehensive picture of what our data looked like and assisted us in identifying patterns. We did this by constructing frequency and percent distributions. We used Descriptive Statistics in order to “describe” the data set. The most common descriptives used are: mean, minimum and maximum values, median, and mode. 3.4.2 Analysis of the Qualitative Data According to Lichtman (2013, p. 257) in relation to phenomenological analysis, he said: “If you were following a phenomenological tradition, you would be interested in the lived experiences of the individuals. You will need to explore some of the philosophical underpinnings of phenomenology, and your data analysis would be facilitated if you bracketed your views. You might choose to conduct either narrative analysis or thematic analysis.”

23

Therefore, in order to analyse the qualitative data, we chose to follow the six steps recommended by Lichtman (2013):

Step 1. Initial coding. Going from responses to summary ideas of the responses. Step 2. Revisiting initial coding. Step 3. Developing an initial list of categories. Step 4. Modifying initial list based on additional rereading. Step 5. Revisiting your categories and subcategories. Step 6. Moving from categories to concepts.

All the 13 interviews were audio recorded and transcribed into text later on. The transcriptions were done verbatim, as literally as we perceived them from the interviewees. When we got all the interviews in text we proceeded to code them. Once we coded them we revisited that initial coding, we excluded the irrelevant and repeated codes (maybe different names but same meanings) and we thought on categories. After having the categories, we created a hierarchy with categories and subcategories. Finally, we selected from those categories and subcategories the main concepts related to our research questions. 3.5 Reliability, Validity or similar Yin (2003) proposed four criteria that permit to the researcher to test the scientific quality of his/her research. These four criteria are: construct validity, internal and external validity, and the reliability of the research. Construct Validity Construct validity refers to establish “correct operational measures for the concepts that are studied” (Yin, 2003). The level of construct validity is improved by using multiple sources of evidence and establishing a chain of evidence. In our case, the theoretical data came from books, articles, and research reports, while our empirical one came from the surveys and the interviews. This demonstrates that we used multiple sources in order not to be subjective with the data we got. Internal Validity Internal validity means “establishing a causal relationship, whereby certain conditions are shown to lead to other conditions” (Yin, 2003). According to the author, internal validity is only important for explanatory case studies where causal relationships are observed. Our thesis, however, is mainly exploratory and does not treat with causal relationships, so the criterion of internal validity is not important for us. External Validity External validity deals with “the domain to which a study’s findings can be generalized” (Yin, 3003). So, the theory has to be tested in some different settings and the findings have to be the same. Due to the fact that we got more than 100 surveys, we had the possibility to generalize if the findings were similar in the majority of them. Reliability The reliability test comes to demonstrate “that the operations of the study – such as the data collection procedures – can be repeated with the same results”. Hence, the objective

24

is to minimize the study’s biases and errors. Our study research was based on a survey among the public and several interviews, so our interpretations and conclusions also rely on these people’s answers, which can be subjective. We reduced this flaw by asking a larger amount of people for the survey, for instance. In the case of the interviews, they were recorded and transcribed literally word for word. Since we got two different sources of information, the interviews (qualitative) and the questionnaires (quantitative), our plan was to triangulate the data in order to identify themes. These themes were taken back to the participants so they could confirm if they felt that the themes represented what they said. 3.6 Ethical Considerations The ethical considerations recommended by Creswell (2009) that were taken into account during this research were: Disclosure: All participants were informed of the nature and purpose of the research. Furthermore, the confidentiality and anonymity were ensured to all of them. Voluntariness: The participants gave their consent voluntarily and freely of any coercion to participate in the research. Consent: All participants signed up a form prepared to give their consent (See Appendix C.)

25

4. Empirical Findings _____________________________________________________________________________ This chapter presents the empirical findings from our research. First, we present the empirical evidences from the quantitative part of our study (presentation and interpretation of statistical analyses) and then the empirical evidences from the qualitative part of our study (interview excerpts and their interpretation). _____________________________________________________________________________ 4.1 Evidences from the quantitative part These evidences are based on our main research method that was based on the survey. We got a total of 128 respondents who filled out our survey in any of the three different ways we offered: online, on paper and through the phone. The first initial data that we wanted to gather from our respondents was their ages. Having the ages was important because with them we could distribute our respondents between the two age-groups, 16 to 35 and 36 to 70. In the following histogram (Figure 6) we represented the ages of them in the axis X and their frequency in the axis Y. The interval of our sample went from 16 years to 66 years old.

The histogram shows the absolute numbers of the respondents for each age. We chose this type of diagram because it presents clearly the type of information we wanted to show. We can see, for example, that we got 13 respondents who were 23 years old and the same number of respondents were 25 years old. Furthermore, we can perceive that our youngest respondent was 16 and the oldest one was 66. The group 1 (people between 16 and 35) got more respondents than the group 2 (from 36 to 70). The highest population in the group 1 happens in the interval between 21 and 29 years old, and for the group 2 it happens in the interval between 36 and 42. After adding the frequency for each age in the interval of 16-35, respectively in interval of 36-70, we got the results in the table below (Table 2):

1 1 0 1 253

13

4

13

54 3

6

0 453 3 1

54 4

56

2 3 4 0 1 3 4 1 0 1 2 1 1 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 002468101214

16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70

Freq

uency

Agesofrespondents

Histogram

Frequency

Figure 6: Histogram that represents the age of the respondents in absolute numbers.

26

Group 1 (16-35) Group 2 (36-70)

77 respondents 60.2% 51 respondents 39.8% As we previously stated, we tried to get equal number of respondents in both groups in order to make the results of our research more accurate. At the end we got around 60% of respondents from the first group and around 40% in the second. In the survey, our second question was if the respondent was a Facebook user or not. In the case of the online survey we got 100% Facebook users as well as with the targeted phone calls. However, we got 9 respondents who did not use Facebook (mostly from the group 2) while we delivered the survey on paper. Therefore, the number of Facebook users in the group 1 decreased by one respondent and the group 2 decreased by eight respondents. At the end, the new table with only Facebook users was (Table 3):

Group 1 (16-35) Group 2 (36-70) 76 respondents 63.9% 43 respondents 36.1%

Since our study was about studying the users’ perceptions of privacy on Facebook, it was an obvious decision to use the data from Table 3 (only Facebook users) for our analysis. In total we had 119 respondents who used Facebook of which 76 belonged to the group 1 and 43 to the group 2. In relation to the privacy itself, our respondents shared their publications on Facebook in different ways: some chose to share publicly, others to all their friends, others to just a few friends and the last ones only with themselves. We showed this in the next Table 4:

Public All friends Few friends Only you 21 84 12 2 18% 71% 10% 0.1 %

As we can see, most of our respondents shared their posts just with their friends. We thought that it could be more informative for our research to compare the answers from both groups (Table 5):

Public All friends Few friends Only you Group 1 (16-35) 12 (16%) 56 (74%) 7 (9%) 1 (1%) Group 2 (36-70) 9 (21%) 28 (65%) 5 (12%) 1 (2%)

Table 2: Percentage distribution of the two groups.

Table 3: Percentage distribution of the two groups (only Facebook users).

Table 4: Users privacy levels on Facebook and their percentage.

Table 5: Comparison of the groups’ posts privacy levels on Facebook

27

Using 3D bar diagram to represent the absolute numbers, we got (Figure 7):

In both groups, most of the respondents shared their posts just to their friends. Based on Table 5 and Figure 7 we got that:

1) Respondents from group 1 share their posts with their friends slightly more than respondents from group 2.

2) Respondents from group 2 share their posts publicly slightly more than respondents from group 1.

Also related to the privacy, we asked our interviewees if they had ever read the privacy policies on Facebook. We got the following results (Table 6):

YES NO Group 1 38 (50%) 38 (50%) Group 2 22 (51%) 21 (49%)

In this case, we got almost the same results comparatively in both groups. In the next question we asked our respondents how they felt about the statement: “I think I share too much private information”. In order to answer this, we offered a five point Likert’s scale and we got the following results (Figure 8):

0

10

20

30

40

50

60

allfriends public justafewfriends justyou

56

127

1

28

95

1NUMBE

ROFPA

RTICIPAN

T

Comparisonofprivacyoftheposts

Figure 7: Comparison of the groups’ privacy levels on Facebook (blue = Group 1; red = Group 2)

Table 6: Respondents who have read the privacy policies on Facebook

28

In the question about understanding the main concept of privacy on Facebook we got the results shown in the following Table 7:

1 - strongly disagree

2 - somewhat disagree

3 - neither agree nor disagree

4 - somewhat agree

5 - strongly agree

Group 1 0 (0%) 8 (11%) 22 (29%) 34 (45%) 12 (16%) Group 2 5 (12%) 5 (12%) 9 (21%) 15 (35%) 9 (21%)

We could see here that there was a relevant number of respondents in the group 2 who strongly disagreed with the statement, what meant that they did not feel like they understood the main concept of privacy on Facebook. With respect to the question of how many of our respondents had changed their privacy settings on Facebook, we got that 97 of them (81.5%) had done it. Compared into the two groups, Table 8:

YES NO Group 1 66 (87%) 10 (13%) Group 2 31 (72%) 12 (28%)

Here we could see that a bigger percentage of respondents from group 2 did not change privacy settings on Facebook in comparison with group 1.

051015202530

STRONGLYDISAGREE

SOMEWHATDISAGREE

NEITHERAGREENORDISAGREE

SOMEWHATAGREE

STRONGLYAGREE

1 2 3 4 5

2621

1613

0

15 14

74 3

ComparisonaboutsharingtoomuchprivateinformationonFacebook

Figure 8: Opinions about sharing too much private information on Facebook (blue = Group 1; red = Group 2)

Table 7: Comparison of understanding the main concept of privacy on Facebook

Table 8: Comparison of changing the privacy settings on Facebook

29

When we asked the respondents if sharing on Facebook affected their privacy, we got that 79 of them agreed (66.3%). Compared into the two groups, we got Table 9:

YES NO Group 1 54 (71%) 22 (39%) Group 2 25 (58%) 18 (42%)

A higher percentage of respondents in the group 1 agreed that sharing affects their privacy on Facebook. Then we asked the respondents how they felt about the statement: “You lose your privacy by using Facebook”, and we got the results on Table 10 and Figure 9:

1 - strongly disagree

2 - somewhat disagree

3 - neither agree nor disagree

4 - somewhat agree

5 - strongly agree

Group 1 7 (9%) 18 (24%) 21 (28%) 19 (25%) 11 (14%) Group 2 6 (14%) 11 (26%) 11 (26%) 11 (26%) 4 (9%)

A bigger percentage of respondents in group 2 strongly disagreed with the statement that they lose privacy by using Facebook. On the other hand, a bigger percentage of respondents in group 1 strongly agreed that they lose privacy by using Facebook.

0510152025

STRONGLYDISAGREE

SOMEWHATDISAGREE

NEITHERAGREENORDISAGREE

SOMEWHATAGREE

STRONGLYAGREE

1 2 3 4 5

7

1821 19

116

11 11 11

4

ComparisonoflosingprivacybyusingFacebook

Table 9: Comparison of sharing affects privacy on Facebook

Table 10: Comparison of losing your privacy by using Facebook

Figure 9: Comparison of losing your privacy by using Facebook (blue = Group 1; red = Group 2)

30

In the next question from the survey we asked the respondents which privacy settings they chose for different concepts in their profiles such as: their personal information, birthday, education and work, etc. We summarized them in the next two tables, Table 11 and Table 12 (see Appendix D for diagrams of Table 11 and Table 12): For group 1:

Personal Information

Birthday About Me

Education and Work

Religion and

Political View

Relationship Photos and

Videos

Status

EVERYONE 12 16 12 21 10 10 13 13 FRIENDS OF FRIENDS

7 9 12 10 7 8 11 12

JUST FRIENDS

46 39 47 38 36 38 51 47

JUST ME 11 12 5 7 23 20 1 4

For group 2:

Personal Information

Birthday About Me

Education and Work

Religion and

Political View

Relationship Photos and

Videos

Status

EVERYONE 7 11 10 16 11 14 9 10 FRIENDS OF FRIENDS

3 3 6 5 3 3 3 6

JUST FRIENDS

24 24 23 19 20 20 29 24

JUST ME 9 5 4 3 9 6 2 3

Taking into account the relative values from both groups, we got: 1) Group 2 shares information about Education and Work more publicly than Group 1. 2) Group 2 shares information about Religion and Political View more publicly than Group 1. 3) Group 2 shares information about Relationship more publicly than Group 1. 4) Group 1 shares information about Birthday with friends of friends more than Group 2. 5) Group 2 shares information about About Me more publicly than Group 1. 6) Group 2 shares Photos and Videos more publicly than Group 1. 7) Group 1 shares Photos and Videos with friends of friends more than Group 2.

Table 11: Different privacy settings for private information within the group 1.

Table 12: Different privacy settings for private information within the group 2.

31

4.2 Evidences from the qualitative part From our interviews and using the method of the three C’s: Coding, Categorizing, and Identifying Concepts, we got a first table (Table 13):

Interview keywords

Initial coding framework

Concepts

Group 1 Group 2 • 29 • 32 • 19 • 31 • 27 • 28 • 28

• 51 • 59 • 36 • 66 • 50 • 45

• Perception of ages Age

• 6 years • 7 years • 2,5 years • 10 years • 8 years • Like 10 years • About 10 years

• 7 years • Around 5 years • 11 years • 10 years + • More than 10 years • From beginning of

implementation of Facebook

• Perception of the Facebook usage time period

Period

• Daily • Daily • Daily • Daily • Daily (few times per

day) • Every day (few times) • Every day

• Daily • Daily (once a day) • Daily • Every day • Few time on day, but

less and less • Every day

• Perception of the frequency of use of the Facebook

Frequency

The concept of Age was important for us because our research was about the comparison of certain attitudes among people in different ages. With this concept, we wanted to show the age of the participants who we had our interviews with. By introducing the concepts of Period and Frequency, we wanted to present that our respondents were users of Facebook for a long period and that they used it everyday. The idea behind these two concepts was to show that we chose respondents with a long run of use of Facebook and that Facebook was part of their everyday lives. By choosing these respondents, we felt that we had more reliable information than interviewing those who used Facebook just several times a month or new users. In the next table (Table 14) we summarized the concepts we found related to privacy during the interviews with the respondents of group 1: (for an extended version of the table including keywords from the interviews, check Appendix E)

Table 13: Concepts of Age, Period and Frequency.

32

Initial coding framework Concepts • Knowledge about privacy • Nescience about privacy

Knowledge about privacy

• Changed privacy settings • Not changed privacy settings

Changing privacy settings

• Friends • Friends of friends

Set up the privacy settings

• Perception of reading privacy policy • Too much text • Privacy settings types • Private level

Reading privacy policy

• Perception of how sharing on Facebook affect your privacy

• Worried/Not worried • Thinking about consequences • Existing/not existing privacy

Sharing on Facebook affect your privacy

• Perception of identifying risk • Negative notions about risks • Recognize/Not recognize risk • Tracking Swish • Fake profile • Stealing identity • Robbery

Identifying the risk of sharing private information on Facebook

• Perception of the influence of knowledge to privacy risks

• Information use against one • Double authentication • Became more careful what share • Limit to integrity

Knowledge about privacy risks

In the next table (Table 15) we summarized the concepts we found related to privacy during the interviews with the respondents of group 2: (for an extended version of the table including keywords from the interviews, check Appendix E)

Table 14: Concepts related to Privacy within group 1.

33

Initial coding Concepts • Knowledge about privacy • Nescience about privacy

Knowledge about privacy

• Changed privacy settings • Not changed privacy settings • Nothing to hide • Tagging option • Hide things about self • Better security • Avoid unknowing people

Changing privacy settings

• Just friends • Only me • Public • Combination

Set up the privacy settings

• Perception of reading privacy policy • Check if Facebook is secure • Post something public • Avoid contact from unknown people • Not get advertisement • Who can use your private information

Reading privacy policy

• Perception of how sharing on Facebook affect your privacy

• Information use against one • Think before share • Spread private information • Take inloggning details • Possibility that your privacy is looked by

other person • “No life changing” share • Protect yourself

Sharing on Facebook affect your privacy

• Perception of identifying risk • Negative notions about risks • Recognize/Not recognize risk • Just share info about dogs • Private information to be sold in the

future • Identity theft • Private information for advertising

Identifying the risk of sharing private information on Facebook

• Perception of the influence of knowledge to privacy risks

• Risk when you share private information • Protect yourself • Consider what write • Adapting privacy settings • Not giving information to third parties

Knowledge about privacy risks affect to use of Facebook

Table 15: Concepts related to Privacy within group 2.

34

The knowledge about privacy was an interesting concept for our research because we wanted to explore the perceptions of the different age-groups about privacy. The next concept we found was “Changing privacy settings”. Through this concept, we wanted to find out which group was more inclined to change the privacy settings. However, the answer we got is equal for both groups. The concept of “Set up the privacy settings” represented how the respondents within the two age-groups shared information from their Facebook profiles. The concept “Reading privacy policy” existed because we wanted to find out which group was more aware of the Privacy Policy on Facebook. The concept “Sharing on Facebook affects your privacy” represented which group contemplated more before they published something on Facebook. The concept “Identifying the risk of sharing private information” on Facebook was here to see if the respondents saw the existence of any risk and which risks existed. In both groups, around 50% of respondents saw the existence of a risk, while the other half did not recognize any risk. The last concept was “Knowledge about privacy risks” where we wanted to find out how risk knowledge affected the disclosure of private information and to what extent the group had a greater amount of risk knowledge. Based on this analysis we got following findings: 1) Respondents from group 2 (36-70) had more knowledge about privacy on Facebook than respondents from group 1 (16-35). 2) A bigger number of respondents from group 2 (36-70) read the Privacy Policy on Facebook than respondents from group 1 (16-35). 3) Respondents from group 2 (36-70) agreed more with the statement that “Sharing on Facebook affects your privacy” than respondents from group 1 (16-35).

35

5. Discussion ______________________________________________________________________ In this chapter we discuss the findings from the previous chapter. We start with a summary of the main concepts related to the research questions and then we continue with our own reflections compared with the theory. ______________________________________________________________________ 5.1 Findings Discussion The main idea of this study was to research about how different age-groups cared about their privacy on Facebook. We wanted to know if their privacy considerations were the same or there existed some differences. Lastly, we wanted to check if the existence of privacy risks affected the use of Facebook between the two different age-groups. These topics were discussed by the RQ1 and the sub questions SQ1 and SQ2. In the next subchapter we discussed the empirical findings we got from the previous chapter. 5.1.1 Discussion of Research Question Since our research was based on the mixed methodology, we got different kind of data: numerical/statistical from quantitative, and personal thoughts and experiences from the qualitative part of study. The questions in the survey (quantitative) and the questions in the questionnaire (qualitative) were almost the same because the important thing for us was the different type of data from the answers. The idea here was to see if the different methodologies provided of the same results for our research question. Using the concurrent triangulation approach, qualitative and qualitative data were collected simultaneously and then analysed individually. At the end, the findings were compared by transforming one type of data into the other type to determine “the extent to which the analyses confirm or contradict one another” (Creswell, 2009, p.213). The concurrent triangulation design of this study contributed with the same importance to both qualitative method and quantitative method, and the data derived from each method. We reported firstly the findings based on the analysis of quantitative data and then we reported the findings extracted from qualitative data. The order in which the findings were reported did not indicate order of importance. In order to answer the main research question, we firstly started answering simpler questions. The key word from the research question is "care" (How do young adults and adults care about their privacy on Facebook?). The term "care" describes protection from something or someone, doing something right, as well as avoiding possible threats or risks. In our case, we talked about how users protected their privacy on Facebook, whether they had any privacy concerns about it, if they had changed their privacy settings to avoid compromising their privacy, or even understanding the main concept of privacy in this SNS. 5.1.1.1. Quantitative discussion After the analysis process, we mentioned that most respondents, regardless of the age group, shared their posts with all their friends. Since the purpose of our research was to determine which age group and how they took care of privacy on Facebook, we went

36

deeper in the analysis of the obtained data and we found out that group 1 shared more posts with all their friends, while group 2 was more likely to share publicly. Furthermore, in both age groups, an equal number of respondents read and did not read the Privacy Policy on Facebook. As far as sharing too much information on Facebook was concerned, a certain number of respondents in group 2 strongly agreed, while no respondents from group 1 offered such response. This finding suggested that respondents from group 1 did not share too much information publicly, while a predetermined number of respondents from group 2, knowingly, shared too much private information. Moreover, continuing with the comparison of these two groups about the understanding of the main privacy concept, we got the following finding: A predetermined number of respondents in group 2 completely did not understand the term privacy, while no respondents from group 1 provided that kind of answer. The next thing we realized was that larger percentages of respondents in group 2 compared to respondents in group 1 had not ever changed their privacy settings. In relation to the statement “Facebook affects your privacy”, respondents from group 2 were strongly disagreeing considerably lower percentage than participants in group 1. We came to the conclusion that respondents from group 2 did not perceive Facebook as a threat to their privacy. Unfortunately, through the quantitative analysis, we did not get the answer to the question why. Related to sharing various types of information on Facebook (photos, about me, education, etc.), we found that most of the respondents in group 2 were more inclined to share private information. Thus, respondents in group 2 shared the following information publicly in relation to respondents in group 1:

Since we only provided numeric values through the survey, the next question to answer our research question was to analyse the qualitative data, which is presented in the next subchapter.

Figure 10: Group 2 shared this private information more publicly than group 1

37

5.1.1.2. Qualitative discussion Through the method of interview, we received "broader" answers based on the personal experiences, knowledge and thoughts of our respondents. We knew that the term of knowledge was not an easily measurable term, but what we wanted to get from our respondents was how they personally evaluated their knowledge about privacy on Facebook. Firstly, we got that respondents from group 2 thought that they knew more about privacy on Facebook than respondents from group 1. In the concept “changing privacy settings”, most respondents answered that they had changed their privacy settings sometime. Related to the same concept, we also received the response to which specific settings they had changed. Some had changed their privacy settings for posts, tags, statuses, photos, etc., while others answered that they had changed their privacy settings for better security or because they wanted to hide some information about themselves. For instance, one of the respondents decided to change her privacy settings after she became a mother. Therefore, she confirmed that she changed her privacy settings over time caused by other circumstances: “Yes, I changed it (privacy setting) to protect the pictures I upload when I got my baby”. She was aware of the risks that existed when sharing private information, no matter if they are pictures or some other private information. In the next concept, “set up privacy settings”, we got diverse answers. While respondents from group 1 mostly responded that their private information was shared only with their friends or friends of friends, respondents from group 2 responded more diversely, from only me, through a combination of settings to public. Furthermore, through the qualitative data analysis, we discovered that a bigger number of respondents of group 2 read the Privacy Policy on Facebook than respondents from group 1. In relation to the next concept, “Sharing on Facebook affects your privacy”, we noticed that respondents from group 2 agreed more with that statement than respondents from group 1. Here, besides the positive or negative response from our respondents, we got different explanations of ways in which Facebook affected their privacy. For example, one respondent from group 2 mentioned the risk of robbery of houses when publishing photos from vacation, fake profiles, spreading private information by another person, etc. On the other hand, one of the young respondents (group 1) stated that he was not concerned about privacy on Facebook, but he “thinks it's more difficult to have privacy if one is on Facebook. If you want to have privacy you should delete Facebook.” Another respondent from group 2 who agreed with the statement provided with the kind of information he shared on Facebook that could not compromise his privacy, namely: “stuffs that are “no life changing”, funny stuffs that are not so much important”. Through the concept “identifying the risk of sharing private information”, we wanted to find out how deeply our respondents thought about the risks and which risks they perceived more often when they shared private information. Most of the young respondents (group 1) did not recognize the risks of sharing private information on Facebook, but one of the respondents said: “There is always a risk when one uses Social Media”. The risks identified by younger respondents were: tracking swish, robbery, fake profiles, while the risks identified by the older respondents were: identity theft, private

38

information for advertising and, as one said, "the information that we share today on Facebook will be sold in some years ahead.” In the concept “knowledge about privacy risks”, one of our respondents from group claimed: “I don’t share that kind of information that any other person can use against me” and another one: “I think when one knows the risks and the privacy on Facebook, one becomes more careful with the type of information one shares”. On the other side, respondents from group 2 considered that they should protect their private information more, not share information such as their place of residence and be careful not to overwhelm their private messages. 5.1.1.3. Comparison of quantitative and qualitative discussion By comparing these two research strategies using concurrent triangulation, we could conclude that they were complementary to each other and thus enabled us to get a more comprehensive answer to our research question. Under the term “reading privacy policy on Facebook”, we got different answers for different methodologies. In the quantitative part we got almost the same percentages of young respondents and old respondents who read the privacy policy, while in qualitative part we got the result that older respondents read privacy policy more than younger respondents. When we talk about how sharing on Facebook affects the privacy of users, from the quantitative part we got that younger respondents agreed with that more than older respondents, while we got opposite results through the qualitative analysis. Regarding to the change of privacy settings, we could see that through both analyses most of the participants had changed their settings in both groups. What was also noticeable was that there was less difference between the groups in the qualitative analysis, but the result was the same in both cases anyway. 5.1.1.4. Answer and conclusion of Research Question In this part, we started answering the sub-questions and after that we continued with answering the main research question. SQ1) Is the privacy consideration the same within the young adults and adults who use Facebook? Through our research, we can respond negative to this question. According to the results of the qualitative analysis, older respondents believed that they had more knowledge about the term of privacy than younger respondents, while older respondents shared more information publicly than younger. We can conclude that no matter how older respondents had more knowledge about privacy, they shared their privacy easier than younger. As far as sharing too much information on Facebook, we came up to contradictory results between quantitative and qualitative analysis. While the younger respondents in the quantitative analysis were more accustomed to this statement, the qualitative analysis gave us opposite results.

39

SQ2) Do the privacy risks on Facebook affect the use of this SNS among the two groups? Through our research, we can respond affirmative to this question. Yes, the privacy risks affected the use of Facebook. Since our research was comparative between the two age groups, our aim was to compare the response between these two groups. However, in this case this was not possible because of the opposite results obtained through quantitative and qualitative analysis. Unfortunately, we were not able to give an answer to which age group in a larger number answered positively. The reason for this was that in the quantitative analysis, younger respondents answered with a higher percentage positively, while in the qualitative analysis, a larger number of older respondents answered positively. Therefore, in order to achieve a clear and unanimous response to this question, we believed that we should have deepen the research and conducted more interviews. In this way, it would have been established whether the reason for the failure to give a unanimous response was the lack of a larger number of respondents through the qualitative method or the result would remain the same as in our case. RQ1) How do young adults and adults care about their privacy on Facebook? Young-adults and adults care about privacy on Facebook on the next ways:

• By changing the initial Facebook privacy settings. • By reading and introducing the Privacy Policy of Facebook. • By identifying possible risks when sharing private information on Facebook. • By understanding the main concept of privacy on Facebook.

Since our research was comparative, what we wanted to show with the response of the main research question is in which amount the different age groups care about the above-mentioned points through the mixed methodology.

• By changing the initial Facebook privacy settings

As we knew, both age groups were concerned about privacy by changing the original privacy settings on Facebook. Through the quantitative method, we got the result of 87% younger respondents who changed their settings and 72% older respondents. Through the qualitative method we obtained similar results, i.e. in both age groups, an equal number of respondents changed their settings, apropos they did not. Furthermore, we can conclude that most of the participants from both age groups changed their settings based on the results obtained through both analyses.

• By reading and introducing the Privacy Policy of Facebook

Another way how Facebook users are concerned about their privacy is by reading Privacy Policy on Facebook and getting to know the key terms from it. In this case, we got different answers through the different analyses. So, in the quantitative analysis, we got the result of 50% of the younger participants who had read it and the 51% of the older ones who had read it, while through the qualitative analysis, we got the result of a significantly higher number of older respondents who had read it more than the younger ones. We can conclude that we have obtained different results through the different types of analysis. Basically, older respondents in both cases had read the Privacy Policy on Facebook more than younger. Only in the qualitative research the differences were more significant than in quantitative where the difference was only 1%.

40

• By identifying possible risks when sharing private information on Facebook

We got the results for this concept from the qualitative analysis because there we had the opportunity to ask additional questions through the interview method, and also the respondents had the opportunity to explain their answers, which is not possible in surveys (quantitative). In the group of younger respondents, the most commonly identified risks were: tracking swish, robbery, and fake profiles; while the risks identified by older respondents were: identity theft, sharing private information to advertisement, and selling our private information in future. We can conclude that both age groups are aware of the existence of the risk, only by noticing the different risks when sharing private information on the Facebook.

• By understanding the main concept of privacy on Facebook

We got the results for answering this expression just from the quantitative analysis and in those results we could see the differences in the responses between the age groups. While younger respondents did not have a single answer that they did not fully understand the main concept of privacy on Facebook, from the older respondents we had, 12% gave this answer. Based on this we can conclude that older respondents are not fully acquainted with the concept of privacy on Facebook. 5.2. Overall Discussion Here in this section we discuss the results we got in this research with previous research conducted in the same topic. In Norway, Brandtzæg et al. (2010) found that younger and older adults displayed completely open public profiles without realizing it. Also, that younger users stated that they used Facebook less, more cautious than before and they were aware of and changed privacy settings compared to the older ones. Through our research, we cannot say much about the rising or the decrease of use of our participants on Facebook because that was outside of our scope, but we can confirm that still younger adults are aware and change the privacy settings more than older ones (See Table 8 on page 28). In Turkey, Kezer et al. (2016) compared three age groups (18-40, 41-65, and 65+) and discovered that although all the age groups were comparable in terms of online privacy literacy, younger groups were more tending to self-disclose and engage in privacy protective behavior on Facebook. Related to privacy attitudes, older age groups were more likely to be worried about privacy of other individuals. Moreover, the impact of privacy attitudes on privacy protective behavior was stronger among mature adults. In our study, we could see that privacy protective behavior in the younger adults in terms of that they shared most just with friends or/and friends of friends (See Table 14 on page 32). In the Netherlands, Steijn et al (2016) confirmed with their study that older individuals were more probable to associate situations related to personal information and reported more concern regarding their privacy. In this case, throughout our research, we found out that both age-groups were equal concern about their privacy. This can be shown, for

41

instance, when we asked if they had read the Privacy Policy on Facebook and around half of both groups responded affirmatively (See table 6 on page 27).

42

6. Conclusion ______________________________________________________________________ In the last chapter of this thesis the conclusions are presented. Then we summarized the results we got in this thesis and our contribution to the Information Systems and privacy fields. We ended up the chapter with suggestions for future research. ______________________________________________________________________ This thesis work reported on a study of a comparison of user’s perception of privacy issues on Facebook between younger adults and adults. It was found that users in the two different age-groups did not have the same privacy consideration in the use of Facebook. Furthermore, privacy risks on Facebook affected at some extent the use of this SNS among these two groups. Lastly, that young-adults and adults care about their privacy on Facebook in terms of:

• By changing the initial Facebook privacy settings • By reading and introducing the Privacy Policy of Facebook • By identifying possible risks when sharing private information on Facebook. • By understanding the main concept of privacy on Facebook

In general, we believe that the results presented in this work increase the knowledge about the rise of new social groups and use by different generations of Facebook. Young adults’ often experience a bigger self-awareness and choose compliance as a way to maintain their social privacy, while adults claim to have more knowledge about it. The results and analysis in this thesis should be seen as limited in the sense that our aim was not to provide a representative picture but rather an in-depth exploration of the social aspects of the social “privacy dilemma”, comparing two samples of younger adults and adults. We realized how difficult this task was due to each individual, no matter which age group he/she belongs to, is different and unique. The research was conducted using the interpretative paradigm, more specifically, the Mixed Methods Phenomenological Research (MMPR). The intention with this was to get an answer to the main research question and the two sub-research questions that were focused on the “privacy dilemma” on Facebook. The resulting data was analysed according to the principles of phenomenology. Care was taken to guarantee that the research was done in an ethical manner in each one of the different stages. 6.1. Contribution We hope that this study has provided more light with the experiences of people in different ages using Facebook and their concerns about privacy risks that there exists. This thesis can be useful for SNSs’ providers such as Facebook, in the sense of that they can see how people perceive the SNS in their different stages of life. This work can also be valid for future researchers in the field who decide to explore a phenomenon using a mixed methods strategy. This research complements previous approaches and from a not very used angle of research.

43

The present research contributes to the theory on privacy on SNSs and can help SNS practitioners and social privacy researchers to promote sociability and content sharing while supporting privacy. 6.2. Future Work The main findings of our research were that there were no significant differences in the use of Facebook between the two age-groups of our research. We suggest for a further development of the topic to look deeply at the cultural differences of the respondents. This could be made by taking into account the background origins of every individual. As we know that Sweden is a very mixed-up society, this background can affect the privacy perceptions in the different stages of life. The different devices that people use in order to access Facebook can be also a variable to study. It might not give the same results if people just login in using always a mobile device or a desktop computer. Furthermore, the use of mobile devices can provide of new privacy risks focused on the location of the user. Another suggestion might be to replicate our research in other countries and compare the results with ours. With this in mind, we could find generalizations or disparities in people’s behaviour. Also, it would be interesting to research the privacy issues using different types of SNSs like Twitter or Youtube, different types of information posting, and different types of unwanted audiences. Moreover, digital literacy is another key factor that can impact privacy concerns on SNSs. Bartsch and Dienlin (2016) found a positive correlation between privacy protective behaviours and privacy literacy. A future work could include this variable in another study and research how digital literacy affects privacy concerns in people of different ages. Before finishing, we wanted to remark here the importance of increasing the privacy care and awareness among all the SNSs’ users. Only more knowledge and more research in the field could improve people lives and protect themselves from the risks out there.

44

7. References - Acquisti, A. and Gross, R. (2006). Imagined communities: awareness, information sharing, and privacy on the Facebook. In the Proceedings of Privacy Enhancing Technology (PET 2006), Cambridge, June 28-30, 2006. - Anderchen, S. & Charvát, M. (2016). User Perspective of Privacy Exposure on Facebook: An Examination of Risks Perception Among University Students in Sweden. (Student paper). Linnéuniversitetet. - Anderson, T. (2013). Research Paradigms: Ontology’s, Epistemologies & Methods. PhD Seminar. Universitat Oberta de Catalunya. - Avgerou, C. (2000). Information Systems: What Sort of Science Is It? Omega 28. - Bartsch, M., & Dienlin, T. (2016). Control your Facebook: An analysis of online pri- vacy literacy. Computers in Human Behavior, 56, 147e154 - Bornoe, N., & Barkhuus, L. (2011). Privacy Management in a Connected World: Students’ Perception of Facebook Privacy Settings. Paper presented at Workshop on Collaborative Privacy Practices in Social Media. : Part of the 2011 ACM conference on Computer supported cooperative work (CSCW '11), Hangzhou, China. - Boyd, D. and Hargittai, E. (2010). Facebook privacy settings: Who cares? First Monday, Volume 15, Number 8 - 2 August 2010 http://journals.uic.edu/ojs/index.php/fm/article/view/3086/2589 - Brandtzæg, P. B., Lüders, M., & Skjetne, J. H. (2010). Too many Facebook “friends”? Content sharing and sociability versus the need for privacy in social network sites. International Journal of Human-Computer Interaction, 26, 1006–1030. - Buckman, R. (2005). Too much information? Colleges fear student postings on popular Facebook site could pose security risks. The Wall Street Journal, P.B1. - Cavoukian, A. (2007). When online gets out of line. Information and Privacy Commissioner of Ontario. Fetched 2017-05-02, from: https://www.mun.ca/iap/general/When_Online_Gets_Out_of_Line.pdf - Chang, P. F., Choi, Y. H., Bazarova, N. N., & Löckenhoff, C. E. (2015). Age differences in online social networking: Extending socioemotional selectivity theory to social network sites. Journal of Broadcasting & Electronic Media, 59, 221–239. - Chen, W., and Hirschheim, R. (2004). A Paradigmatic and Methodological Examination of Information Systems Research from 1991 to 2001. Information Systems Journal (14), pp. 197-235. - Chua, W. F. (1986). Radical Developments in Accounting Thought. The Accounting Review (61), pp. 601-632. - Council of Europe (2017). Convention for the Protection of Human Rights and

45

Fundamental Freedoms. Fetched 2017-05-02, from: http://www.coe.int/en/web/conventions/full-list/-/conventions/treaty/005 - Creswell, J.W. (2009). Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. Third ed. Sage, Thousand Oaks, CA - Danezis, G., Domingo-Ferrer, J., Hansen, M., Hoepman, J., Le Métayer, D., Tirtea, R., Schiffner, S. (2014). Privacy and Data Protection by Design – from policy to engineering. - Debatin, B., Lovejoy, J. P., Horn, A.-K. and Hughes, B. N. (2009). Facebook and Online Privacy: Attitudes, Behaviors, and Unintended Consequences. Journal of Computer-Mediated Communication, 15: 83–108 - Eley, B., & Tilley, S. (2009). Online Marketing Inside Out. Melbourne: SitePoint. - Falconer, D. (2008). A Demographic and Content Survey of Critical Research in Information Systems for the Period 2001- 2005. Communications of the AIS (22:30), pp. 547-568. - Giorgi, A. (2009). The descriptive phenomenological method in psychology. Pittsburgh, PA: Duquesne University Press.� - Gray, D.E. (2004). Doing research in the real world. SAGE Publications. - Griensven, H., Moore, A.P., Hall, V. (2014). Mixed methods research – The best of both worlds? Manual Therapy , Volume 19 , Issue 5 , 367 - 371 - Groenewald, T. (2004). A phenomenological research design illustrated. International Journal of Qualitative Methods, 3(1) - Hart, C. (2005). Doing your Masters dissertations. SAGE Publications Ltd

- Hirschheim, R. and H. K. Klein (1989). Four Paradigms of Information Systems Development. Communications of the ACM 32 (10 ): 1199-1216

- Jones, H., & Soltren, J. H. (2005). Facebook: Threats to privacy. December 14, 2005. Fetched 2017-05-02, from: http://www-swiss.ai.mit.edu/6805/student-papers/fall05-papers/facebook.pdf - Kaplan, B. and Maxwell, J.A. Qualitative Research Methods for Evaluating Computer Information Systems. in Evaluating Health Care Information Systems: Methods and Applications, J.G. Anderson, C.E. Aydin and S.J. Jay (eds.), Sage, Thousand Oaks, CA, 1994, pp. 45-68. - Kezer, M., Sevi, B., Cemalcilar, Z., & Baruh, L. (2016). Age differences in privacy attitudes, literacy and privacy management on Facebook. Cyberpsychology: Journal of Psychosocial Research on Cyberspace.

46

- Kietzmann, J. H., Hermkens, K., McCarthy, I. P., and Silvestre, B. S. (2011). Social media? Get serious! Understanding the functional building blocks of social media. Business Horizons, 54(3), pp. 241-251. - Kirkpatrick, D. (2010). The Facebook effect: The inside story of the company that is connecting the world. New York: Simon & Schuster. - Klein, H. K., and Myers, M. D. (1999). A Set of Principles for Conducting and Evaluating Interpretive Field Studies in Information Systems. MIS Quarterly (23:1), pp. 67-93.

- Krasnova, H., and Veltri, N. F. (2010). Privacy Calculus on Social Networking Sites: Explorative Evidence from Germany and USA. In 2010 43rd Hawaii International Conference on System Sciences (HICSS) (pp. 1–10). Presented at the 2010 43rd Hawaii International Conference on System Sciences (HICSS). - Lampe, C., Ellison, N., Steinfield, C. (2006) A face(book) in the crowd: social searching vs. social browsing. In the Proceedings of the Conference on Computer Supported Cooperative Work, Banff, Alberta, Canada, 2006. - Lenhart, A., Madden, M., Rankkin McGill, A., and Smith, A. (2007) Teens and Social Media. Pew Internet & American Life Project. - Lichtman, M., 2013. Qualitative Research In Education: A User's Guide. 3rd ed. Thousand Oaks: SAGE Publications. - Lipford, H.R., Besmer, A., and Watson, J. (2008). Understanding privacy settings in facebook with an audience view. In Proceedings of the 1st Conference on Usability, Psychology, and Security (UPSEC'08), Elizabeth Churchill and Rachna Dhamija (Eds.). USENIX Association, Berkeley, CA, USA, Article 2, 8 pages. - Liu, Y., Gummadi, K.P., Krishnamurthy B., and Mislove A. (2011). Analyzing Facebook privacy settings: User expectations vs. reality. IMC ’11: Proceedings of the 2011 ACM SIGCOMM Conference on Internet Measurement Conference, pp. 61–70. - Madden, M. (2010). Older adults and social media. Washington, DC: Pew Research Center's Internet & American Life Project. - Malik, A., Hiekkanen, K., and Nieminen M. (2016). Privacy and trust in Facebook photo sharing: age and gender differences. Program, Vol. 50 Iss 4 pp. 462 - 480� - McAndrew, F. T., & Jeong, H. S. (2012). Who does what on Facebook? Age, sex, and relationship status as predictors of Facebook use. Computers in Human Behavior, 28, 2359–2365. - McGrath, K. (2005). Doing Critical Research in Information Systems: A Case of Theory and Practice Not Informing Each Other. Information Systems Journal (15), pp. 85-101.

47

- Mengshoel, A.M. (2012). Mixed Methods research—so far easier said than done? Elsevier - Myers, M, D., 1997. Qualitative Research in Information Systems. MIS Quarterly, 21(2), pp. 241-242. - Myers, M. D. (2009). Qualitative Research in Business & Management. London: Sage Publications.� - News Blaze. (2009, january 28). How Social Media Helped Barack Obama to Become the Most Powerful Man. Fetched 2017-05-02, from http://newsblaze.com/usnews/politics/how-social-media-helped-barack-obama-to-become-most-powerful-man_8139/ - Ngwenyama, O. K., and Lee, A. S. (1997). Communication Richness in Electronic Mail: Critical Social Theory and the Contextuality of Meaning. MIS Quarterly (21:2), pp. 145-167. - Orlikowski, W. J., and Baroudi, J. J. (1991). Studying Information Technology in Organizations: Research Approaches and Assumptions. Information Systems Research (2:1), pp. 1-28. - Perrin, A. (2015). 65% of adults now use social networking sites – A nearly tenfold jump in the past decade. Pew Research Center. - Pirly, M. (2015). Facebook use by the less-frequently using demographic: A qualitative study. (Student paper). Linnéuniversitetet. - Raynes-Goldie, K. (2010). Aliases, creeping, and wall cleaning: Understanding privacy in the age of Facebook. First Monday, 15(1). Fetched 2017-05-11, from: http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2775/ 2432 - Richardson, H., and Robinson, B. (2007). The Mysterious Case of the Missing Paradigm: A Review of Critical Information Systems Research 1991-2001. Information Systems Journal (17:3), pp. 251-270. - Selvan, A. (2012). Rules of the road. Modern Healthcare, 01607480, 4/23/2012, Vol.42, Issue 17 - Smith, H. J., Milberg, S. J., & Burke, S. J. (1996). Information Privacy: Measuring Individuals’ Concerns about Organizational Practices. MIS Quarterly, 20(2), 167. - Stahl, B. C., and Brooke, C. (2008). The Contribution of Critical IS Research. Communications of the ACM (51:3), pp. 51-55. - Statista. (2017). Statista – The Statistics Portal. Number of Twitter users worldwide from 2014 – 2020. Fetched 2017-05-02, from https://www.statista.com/statistics/303681/twitter-users-worldwide/

48

- Statista (2017). Statista – Global Social networks ranked by number of users. Fetched 2017-05-02, from: https://www.statista.com/statistics/272014/global-social-networks-ranked-by-number-of-users/ - Steijn, W. M. P., Schouten, A. P., & Vedder, A. H. (2016). Why concern regarding privacy differs: The influence of age and (non-) participation on Facebook. Cyberpsychology: Journal of Psychosocial Research on Cyberspace. - Strater, K., and Lipford, H. R. (2008). Strategies and struggles with privacy in an online social networking community. In Proceedings of the 22nd British HCI Group Annual Conference on People and Computers: Culture, Creativity, Interaction - Volume 1 (BCS-HCI '08), Vol. 1. British Computer Society, Swinton, UK, UK, 111-119. - Stutzman, F., and Kramer-Duffield, J. (2010). Friends only: examining a privacy-enhancing behavior in Facebook. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI '10). ACM, New York, NY, USA, 1553-1562. - Tashakkori, A., and Teddlie, C. (2003). Handbook of Mixed Methods in Social & Behavioral Research. Sage Publications. - Theguardian (2010). Privacy no longer a social norm, says Facebook founder. Fetched 2017-05-02, from: https://www.theguardian.com/technology/2010/jan/11/facebook-privacy - Tsoi, H. K., and Chen, L. (2011). From Privacy Concern to Uses of Social Network Sites: A Cultural Comparison via User Survey. In Privacy, security, risk and trust (passat), 2011 ieee third international conference on and 2011 ieee third international conference on social computing (socialcom) (pp. 457 –464). - Tufekci, Z. (2012). Facebook, youth and privacy in networked publics. In ICWSM 2012 - Proceedings of the 6th International AAAI Conference on Weblogs and Social Media (pp. 338-345) - Universal Declaration of Human Rights (2017). Fetched 2017-05-02, from: http://www.un.org/en/universal-declaration-human-rights/index.html - Van den Broeck, E., Poels, K., & Walrave, M. (2015). Older and wiser? Facebook use, privacy concern, and privacy protection in the life stages of emerging, young, and middle adulthood. Social Media+ Society, 1(2). - Velleman, P.F., and Wilkinson, L. (1993). Nominal, Ordinal, Interval, and Ratio Typologies are Misleading. The American Statistician, Vol. 47, No. 1 (Feb., 1993), pp. 65-72 - Wang, Y., Norice, G., and Cranor, L. F. (2011). Who Is Concerned about What? A Study of American, Chinese and Indian Users’ Privacy Concerns on Social Network Sites. In J. M. McCune, B. Balacheff, A. Perrig, A.-R. Sadeghi, A. Sasse, & Y. Beres (Eds.), Trust and Trustworthy Computing (pp. 146–153). Springer Berlin Heidelberg.

49

- Wang, Y., Leon, P.G., Chen, X., and Komanduri, S. (2013) From Facebook Regrets to Facebook Privacy Nudges. Ohio State Law Journal Vol. 74 p. 1307 - 1335 - Waters, S. and Ackerman, J. (2011). Exploring Privacy Management on Facebook: Motivations and Perceived Consequences of Voluntary Disclosure. Journal of Computer-Mediated Communication, 17: 101–115. - Yin, R. K. (2003). Case Study Research: Design and Methods (3rd ed.). London, Sage Publications - Zephoria. (2017) Zephoria Digital Marketing – The Top 20 Valuable Facebook Statistics. Fetched 2017-05-02, from: https://zephoria.com/top-15-valuable-facebook-statistics/ - Zuckerberg, M. (2009). An open letter from Facebook founder Mark Zuckerberg. Facebook Blog (2 December). Fetched 2017-05-02, from https://www.facebook.com/notes/facebook/an-open-letter-from-facebook-founder-mark-zuckerberg/190423927130/ - Östlund, U., Kidd, L., Wengström, Y., and Rowa-Dewar, N. (2011). Combining Qualitative and Quantitative Research within Mixed Method Research Designs: a methodological review. International journal of nursing studies, 2011

50

Appendices Appendix A:

General introduction by researchers for interview: 1. Thank the participant for accepting to participate in our research. 2. Introduce the participant with the topic of our research. 3. Inform the participant about recording his/her voice for interview. 4. Inform the participant about the questions and the duration of the interview. 5. Inform the participant that he/she has the right not to answer to a question if he/she does not want and that any time has the right to stop the interview. 6. Inform the participant about ensuring the confidentiality of his/her personal information. 7. Introduce the participant with adherence to ethical codes and norms from our side. 8. Researchers and the participant sign the agreement about the interview.

Participant Information Name: Age: Gender: Education: Profession:

Interview questions Initial questions: 1. Do you use Facebook? 2. For how long have you used Facebook? 3. How often do you use Facebook? 4. For what do you use Facebook most? Main questions of interview: 1. How much do you know about privacy on Facebook? 2. What kind of information do you usually share on Facebook? 3. Have you ever changed the privacy settings on Facebook? If your answer is YES, what was the reason for that change? If your answer is NO, why? 4. How do you set up your privacy settings on Facebook? 5. Have you ever read Privacy Concerns on Facebook? Why? 6. Do you concern that your sharing on Facebook can affect your privacy? Why? 7. Why do you share private information on Facebook? 8. Have you recognized any type of risk when you share your private information? Which one? 9. In which way you look on privacy risks when you use Facebook? 10. Did the knowledge about privacy risks affect to your Facebook use? How?

51

11. What will be your ideal privacy settings on Facebook? 12. Regarding on this interview, is there something that you want to add about this topic?

52

Appendix B: This is a research about user privacy on Facebook for our Master Degree Projects in Informatics. Thank you for your participation in our survey and for helping us to get more information in this area. This survey is anonymous and all ethical codes and norms will be applied from our side. If you do not want to answer a question or you want to give up the survey you can do it at any moment. Also, if you do not understand a question or you have any doubt, you can contact us at any time via email: io222bw@student.lnu.se / jromi09@student.lnu.se or contact phone number: 070 584 34 84. Our names are: Irma Obradovac and Juan Tomas Rodriguez

The rules for the survey 1. On every question you can give only one answer. Questions on which you can give more answers will be marked. 2. Answer on a way to circle one of the offered answers. 3. If you circle wrong answer, the same cross out and then circle the correct one. 4. Y means YES, that you agree, and N means NO, you disagree. 5. On question with numbers form 1-5 you answer with just one answer, where:

1 – Strongly disagree 2 – Somewhat disagree

3 – Neither agree nor disagree 4 – Somewhat agree 5 – Strongly agree

6. If you have questions with answers a), b), c), d), just circle one which give the best answer for you.

Participant Information Age: Gender:

Questions 1. Do you use Facebook? Y N 2. Who can see your posts on Facebook?

a) just you; b) just a few friends; c) all your friends;

d) public. 3. Have you ever read Privacy Concerns on Facebook? Y N 4. How do you feel about Facebook’s Privacy Policies?

53

a) No problem; b) Some concern;

c) Concerned; d) Don’t like them at all.

5. You share too much private information on Facebook. 1 2 3 4 5 6. You understand the main concept of privacy on Facebook. 1 2 3 4 5 7. Have you ever change your Privacy Settings on Facebook? Y N 8. Do you concern that your sharing on Facebook can affect your privacy? Y N 9. Do you recognize any type of risk when you share your private information? Y N 10. Does the knowledge about privacy risks affect your Facebook use? Y N 11. You lose your privacy by using Facebook. 1 2 3 4 5 12. Privacy policy is effective on Facebook. 1 2 3 4 5

13. Please review your privacy settings with X and choose your display preference for each of the following types of information on your Facebook account. On this question, you can give more answers.

Everyone Friends Of Friends Only Friends Just Me Personal Information

Birthday About Me

Education and Work Religion and Political View

Relationship Photos and Videos

Status 14. Regarding on this survey, is there anything that you want to add about this topic?

54

Appendix C Hi, our names are Irma Obradovac and Juan Tomas Rodriguez. We work on research for our Master Degree Project and we will be honored that your interview contribute to our research. This is a research about user privacy on Facebook and we want to get answer how different how much people know about their privacy on Facebook and how different ages affect to sharing privacy on Facebook. We are students of Information System Master Programm at Linnaeus University, Växjö, Sweden. If you have any question or dillema, you can contact us at any time via email: io222bw@student.lnu.se / jromi09@student.lnu.se or contact phone number: 070 584 34 84. We are very thankful that you will be part of our research. Before we start with interview, we want to meet you with your rights: 1. Participating in this interview is voluntary. 2. You can stop an interview at any moment without further explanation. 3. If you do not want to answer a question freely accentuate and we will jump to the next one. 4. Your name is strictly confidential. 5. Your answers will be used in our research. 6. Age, gender, education and profession will be used for research purposes as numeric data. Please sign this document if you agree with the previously clause: Participant's signature: Date: Participant's name: (CAPITAL LETTERS) Researcher's signature: Date: Researcher's name: (CAPITAL LETTERS) Please circle one of the offered answers: Do you want get transcription of you interview? Y N Do you like to receive final version of our master thesis? Y N

55

Appendix D

11

12

5

7

23

20

1

4

46

39

47

38

36

38

51

47

7

9

12

10

7

8

11

12

12

16

12

21

10

10

13

13

0 10 20 30 40 50 60

PersonalInformation

Birthday

AboutMe

EducationandWork

ReligionandPoliticalView

Relationship

PhotosandVideos

Status

PrivacysettingsonFacebookforGroup1

EVERYONE FRIENDSOFFRIENDS ONLYFRIENDS JUSTME

0 5 10 15 20 25 30

PersonalInformation

Birthday

AboutMe

EducationandWork

ReligionandPoliticalView

Relationship

PhotosandVideos

Status

9

5

4

3

9

6

2

3

24

24

23

19

20

20

29

24

3

3

6

5

3

3

3

6

7

11

10

16

11

14

9

10

PrivacysettingsonFacebookforGroup2

EVERYONE FRIENDSOFFRIENDS ONLYFRIENDS JUSTME

56

Appendix E

Group 1 (16-35 years old) Interview keywords Initial coding framework Concepts

• Too little • Nothing • Enough • Not very much, just careful with my privacy

settings • A little bit but not very much • Not too much • Knows that on Facebook exist option to set who can

see your stuffs

• Knowledge about privacy • Nescience about privacy

Knowledge about privacy

• Changed settings for profile picture and status • Never changed • Changed • Changed for some posts (share with friends except

some persons) • Limited who can see posts • Changed for protect pictures after became mother • Changed, to be more private and do not want that

anyone be able to get information about her

• Changed privacy settings • Not changed privacy settings

Changing privacy settings

• Just friends => that’s my limit for my integrity • Just my friends • Only my friends • Just my friends • For my friends and their acquaintances (don’t share

information to third parties, like apps, etc.) • Just friends • Only friends

• Friends • Friends of friends

Set up the privacy settings

• No • No • No • No, too much text. • Yes, to know what settings can have and what

privacy settings exist • No, do not care about that • Not reading, but looking for settings about privacy

to keep some stuffs on more private level

• Perception of reading privacy policy

• Too much text • Privacy settings types • Private level

Reading privacy policy

• Not worried about this but I have at the same time respect

• Not think about that • Not feel worried about this • It could happen if I shared a lot of “bullshit”, only

share what I stand for • Do not feel concern, but more difficult to have

privacy if one is on Facebook • Yes, sharing my photos can somebody use for fake

profile • Yes, putting pictures from vacation can cause

robbing of your house, always think what share

• Perception of how sharing on Facebook affect your privacy

• Worried/Not worried • Thinking about consequences • Existing/not existing privacy

Sharing on Facebook affect your privacy

• People can have an opinion of me (positive and negative), little paranoid with sharing Swish, personal number, bank account... because if somobody track Swish, get everything about you

• Do not recognize any risk • Do not recognize any risk, but little worried about

this • Do not recognize, but there is always a risk when

one uses Social Media • Do not recognize any risk because do not share too

much personal information

• Perception of identifying risk • Negative notions about risks • Recognize/Not recognize risk • Tracking Swish • Fake profile • Stealing identity • Robbery

Identifying the risk of sharing private information on Facebook

57

• Yes, fake profile, stealing identity • Robbery • Don’t share that type of information that any other

person can use against me • Do not know • Few things in order to decrease the risk, one of them

is double authentication • A little bit • When one knows the risks and the privacy on

Facebook, one becomes more careful with the type of information one shares

• No • Never put stuffs which she doesn’t want that

everybody see

• Perception of the influence of knowledge to privacy risks

• Information use against one • Double authentication • Became more careful what share • Limit to integrity

Knowledge about privacy risks

Group 2 (36-70 years old) Interview key words Initial coding Concepts • Not very much • Not much • Quite enough • Know much • Keep informed • Quite a lot, interesting in that topic

• Knowledge about privacy • Nescience about privacy

Knowledge about privacy

• Never changed because there is nothing to hide • Only my friends can see what I post • Changed tagging option • Changed for better security • Changed, avoid that people who does not know

him get information about him • Changed, because like to hide some things about

him

• Changed privacy settings • Not changed privacy settings • Nothing to hide • Tagging option • Hide things about self • Better security • Avoid unknowing people

Changing privacy settings

• Public • Just my friends • To only me • Only friends • Just friends can see, block friends of friends • Somethings can only he see, somethings his

friends and something friends of friends, but nothing public

• Just friends • Only me • Public • Combination

Set up the privacy settings

• No • Yes, interesting to know what can happen if post

something public • No, did not feel the need • Yes, to see if Facebook is secure • Many times, do not want that unknown people

contact him and does not want to get advertisement

• Yes, because is interesting who can use his private information

• Perception of reading privacy policy

• Check if Facebook is secure • Post something public • Avoid contact from unknown

people • Not get advertisement • Who can use your private

information

Reading privacy policy

• No, think little bit before sharing • No, do not share anything that anyone else can

use against me • Yes, someone can take your inloggning details or

spread your private information • Yes, possible that other person can see your

privacy • Yes, because share stuffs which are “no life

changing”, funny stuffs which are not so much important

• Wants to protect yourself, not just from companies, but from people which do not know him

• Perception of how sharing on Facebook affect your privacy

• Information use against one • Think before share • Spread private information • Take inloggning details • Possibility that your privacy is

looked by other person • “No life changing“ share • Protect yourself

Sharing on Facebook affect your privacy

58

• No, because what share is just about dogs • No, do not recognize • Yes, the information that we share today on

Facebook will be sold in some years ahead, identity theft

• Identity theft • Avoid to write things about what other people

may have opinion, ex. family situation, anybody can take anything about you for advertising

• No, did not think about that

• Perception of identifying risk • Negative notions about risks • Recognize/Not recognize risk • Just share info about dogs • Private information to be sold in

the future • Identity theft • Private information for

advertising

Identifying the risk of sharing private information on Facebook

• Do not know • If you share public information, it can be a risk:

where you live, how old you are, … • Adapting the privacy settings, not sending too

much private messages • Yes, protect own information more • Not as much, just consider what write • To not give informations to companies or third

parties

• Perception of the influence of knowledge to privacy risks

• Risk when you share private information

• Protect yourself • Consider what write • Adapting privacy settings • Not giving information to third

parties

Knowledge about privacy risks affect to use of Facebook

2